BLM LIBRARY

88046175

U.S. Department of the Interior
Bureau of Land Management

Medford District Office
3040 Biddle Road
Medford, Oregon 97504

January 1996

People and the
Riven

A History of the Human
Occupation of the IVIiddle
Course of the Rogue River
of Southwestern Oregon

Volume I

Kay Atwood, Dennis J, Grey

I?

S82
.K6
A89
1995

V X

As the Nation's principal conservation agency, the Department of the Interior has responsibility for most of our nationally owned public lands and
natural resources. This includes fostering the wisest use of our land and water resources, protecting our fish and wildlife, preserving the
environmental and cultural values of our national parks and historical places, and providing for the enjoyment of life through outdoor recreation.
The Department assesses our energy and mineral resources and works to assure that their development is in the best interest of all our people.
The Department also has a major responsibility for American Indian reservation communities and for people who live in Island Territories under
U.S. administration.

<^

■9

it#

>f^ C7a Jo BLM/OR/WA/PL-96/004+1792

Front Cover: The picture is of Adeline Billings, a Karulc Indian woman who was the wife of one of the first miners and settlers
in this area: she represents the bridge between the two cultures.

PEOPLE AND THE RIVER:

A HISTORY OF THE HUMAN OCCUPATION OF
THE MIDDLE COURSE OF THE ROGUE RIVER OF SOUTHWESTERN OREGON

Volume I

Prepared for

USDI -Bureau of Land Management

Grants Pass Resource Area

Medford, OR 97504

Contract # 1422H110-P4-5080

Kay Atwood
Dennis J. Gray

TABLE OF CONTENTS

List of Figures i

Illustrations ii

Site Reference Index iii

Acknowledgements iv

Executive Summary v

PART I: The Indians and the River by Dennis Gray

I . INTRODUCTION 1

II. ENVIRONMENT

Introduction 3

Topography and Vegetation 3

Geology 4

Fauna 4

Climate 5

Floods 6

III. ARCHAEOLOGICAL METHODS, THEORIES, AND ISSUES

Introduction 9

Archaeological Methods 9

Regional Cultural and Chronological Models 12

Issues in Regional Archaeology 19

IV. THE ETHNOGRAPHIC PERIOD

Introduction 21

The Takelma 21

The Athapaskans 27

V. NATIVE AMERICANS AND THE RIVER

Introduction 31

Paleo-Indians (12,000 BP to 10,500 BP) 33

The Early Archaic (10,500 BP to 7,000 BP) 34

The Middle Archaic (7,000 BP to 2,000 BP) 42

The Late Archaic (2,000 BP to 250 BP) 48

The Ethnographic Era (2 50 BP to 150 BP) 56

War and Diaspora (1850 AD to the Present) 61

PART II: The Historic Era by Kay Atwood

VI. APPLEGATE RIVER TO GRAVE CREEK

Exploration 68

Mining 7 0

Settlement 80

Tourism 88

VII. GRAVE CREEK TO MULE CREEK

Mining 95

Settlement 113

VIII. POST WAR AND RECENT YEARS (1945-1995)

Introduction 124

Opening the River 124

Flood and Storm 127

Government Management of the River 12 8

BIBLIOGRAPHY 133

LIST OF FIGURES

Figure 1 . Pro j ect Location Map 2

Figure 2. Chronological Periods in Southwestern Oregon 18

Figure 3. Territorial Boundaries in Southwestern Oregon.... 28

Figure 4 . Location of the Excavated Archaeological

Sites within the Project Boundaries 32

Figure 5. Idealized Profile of the Marial Stratigraphy 36

Figure 6. Map of the Rogue River from the Applegate

River to Hellgate Canyon 67

Figure 7. Map of the Rogue River from Hellgate Canyon

to Grave Creek 74

Figure 8 . Map of the Rogue River from Grave Creek to

Horseshoe Bend 94

Figure 9 . Map of the Rogue River from Horseshoe Bend

to Mule Creek 112

ILLUSTRATIONS

Frances Johnson, 19 06 22

Dipnetting for Salmon 24

Housepit Village 25

Canoe on the Rogue 29

Paleo Projectile Point Base from Stratton Creek 33

Excavation at Marial , 1983 35

Excavation at Stratton Creek, 1992 39

Incised Shale from Stratton Creek 43

Projectile Points from Marial 45

Dart points from Marthaller Site 52

Early Projectile Points from Marthaller Site 52

Net-Sinkers from Marthaller Site 53

Stone Bowl Fragment from Marthaller Site 53

Galice, ca. 1910 71

Hydraulic Mining, ca. 1900 77

Almeda Mine, ca. 1910 78

General Land Office Map, 1854 (T 36S, R 6W, W.M) 80

Rogue River in Flood 82

Metskers Map, 1932, T 35S, R 7W, W.M 86

Rogue River in Flood at Almeda Mine, 1927 88

"Speed's Place" on the Rogue, 1987 90

CCC Camp at Rand, Aerial View, ca. 193 5 91

Bridge over Rogue River at Grave Creek 92

Grave Creek Bridge Plaque, 1935 93

Cabin near Rogue River, ca. 1930 9 6

Chinese Miner 9 8

Whisky Creek Cabin, Notched Logs 101

Whisky Creek Cabin, ca. 1925 102

Red River Gold Mining Company Trestle, ca. 1912 103

George Billings at Mule Mountain 1930 105

Unidentified Miners, Rogue River Canyon 106

Red Keller, 1976 107

Price Copsey at Cabin, ca. 1915 108

Lou Martin, 1976 109

Miners at Mule Creek, ca. 1935 110

Walter and May Criteser 113

Rogue River Betweeen Horseshoe Bend and Winkle Bar 114

George, Sarah and Ivin Billings, ca. 1900 115

Adeline Billings, ca. 1915 116

Baskets Made by Adeline Billings 117

Pack Train, Mule Creek 118

George Billings Homestead, 1910 119

Billings Family and Friends, Tabernacle," ca. 1915 120

U.S. Forest Service Crew, ca. 1920 122

Glen Wooldridge 125

Blossom Bar 12 6

BLM Staff with Recreation sign, ca. 1960 128

Cabin in Rogue River Canyon 129

BLM Staff with Rogue River Trail Sign 131

ii

SITE REFERENCE INDEX

Marial

Early Archaic 34,37-3 8

Middle Archaic 43-45

Late Archaic 54

Stratton Creek

Paleo-Indian 33

Early Archaic 38-40

Middle Archaic 46

Late Archaic 55

Marthaller Site

Early Archaic 51-52

Middle Archaic 51-52

Late Archaic 51-52

Ethnographic Era. ...» 53, 57-58

Ritsch Site

Late Archaic 48-50

Ethnographic Era 57-58

Indian Mary Park 75

Flanagan Mine 7 6

Robert Dean Placer Mine 7 6-77

Almeda Mine 78-80

Speed's Place on the Rogue 89-91

Whisky Creek Cabin 101-102

Rogue River Ranch 115-121

m

ACKNOWLEDGEMENTS

The authors would like to thank Jim Leffmann, Eric Schoblom,
Mike Walker, and Becky Brown of the Medford District Bureau of
Land Management's River Program for their support and assistance
during the course of this project. A special thanks to Becky for
her comments on the draft version of this report. Thanks also to
David Brauner of Oregon State University for reviewing an earlier
draft. Kathryn Winthrop, District archaeologist and Contracting
Officer's Representative for this project, patiently reviewed
earlier drafts, provided information and documents, and ably
guided the project from its inception; thanks Kate. We would
also like to recognize the contribution of numerous landowners
within the project boundaries and other interested individuals
who shared with us their knowledge of the land, the people, and
the past.

IV

EXECUTIVE SUMMARY

PART I : The Indians and the River

Chapter II: Environment: The geology, topography, vegetation,
floods, and especially the fisheries of the Rogue River have long
influenced where humans have lived and the how they made their
livings. Native Americans made homes and found sustenance on the
broad terraces between the mouth of the Applegate River and Grave
Creek. Falls and riffles provided good fishing at these
locations, and there was easy access to staple vegetable crops in
the valleys and foothills along the river. Further downriver, in
the canyon below Grave Creek, only the occasional river terraces
allowed them sufficient space and access to resources to maintain
encampments .

For the Euro-American settlers, the broad terraces of the
project area below the Applegate River provided the fertile soils
and access to markets that were necessary for an agricultural
economy. These same locales today still sustain agricultural
activity and, increasingly, residential development. The mineral
wealth of the river corridor drew miners to the region at an
early date. Miners were less in need of open space than farmers;
they settled where they could be near the minerals they mined.
Small benches, rocky bars, or high terraces provided the
necessary settlement prerequisites for the itinerant miner, as
well as for the later corporate enterprises. For native and
settler alike, the periodic floods provided reminders of the
rivers force and the transitory nature of human endeavors.

Chapter III: Archaeological Methods, Theories, Issues:

Archaeology is the primary means of learning about the Indian
history of the river corridor. Archaeologists have worked in this
area for over twenty years, defining the local chronology and
those issues which guide research. Sites along the river indicate
human presence for at least 8,000 years, and possibly longer.

Within this long time period, the changes which occurred in
southwestern Oregon Indian societies are a matter of on-going
debate. The archaeological record suggests that there was both
long-term cultural stability, as well as significant change.
Small, mobile groups of people used the rivers' resources for
thousands of years. At some point in time cultural practices
changed, and a less mobile, more sedentary way of life emerged.

Today, there are significant research issues which guide
archaeological work along the corridor. These issues include
further describing the earlier and later ways of life, as well as
explaining the causes for cultural change; understanding past
climate and environmental changes and their effects on the people

who lived here; documenting the migration of new peoples into the
area, especially those who came from the north, speaking
Athapascan languages; and the resolution of technical issues
specific to the area, such as the utility of stylistic dating.

Chapter IV: The Ethnographic Period. Based on accounts of native
survivors, early anthropologists have sketchily described the
Indian way of life at the time of contact with Euro-American
settlers, in about 1850. With some caution, archaeologists can
use these cultural descriptions to help explain the
archaeological record.

At the time of contact, the Takelma, a Penutian- speaking
people, resided in a territory that centered on the upper and
middle stretches of the Rogue River. The principal villages of
the Lowland Takelma were centered along the Rogue River,
extending from the present-day town of Gold Hill downriver to
perhaps Grave Creek. Athapascan- speaking peoples also inhabited
portions of the river corridor and its tributaries, including the
Applegate River and Galice Creek. Despite differences in
language, the Takelma and the Athapascan groups shared many
similarities in terms of material culture (i.e. tools, clothing,
architecture, etc.) and subsistence economy.

The Takelma relied upon acorns, camas (a root crop) , deer
and elk, and fish, as well as a host of other plants and animals
for food and materials. During the winter, they lived in
permanent villages located at low elevations near rivers and
streams. During the other months of the year, they would move to
temporary camps in the surrounding uplands to hunt and to gather
plants. The local village was the principal social and political
unit, and wealth helped determine a person's status within the
group. The religious and ceremonial aspects of life were not
easily separated from daily tasks, and there was no great
separation between the spiritual and the secular.

Chapter V: Native Americans and the River: Several major
archeaological sites along the Rogue River provide evidence of
the long history of the Indian people in this region. This
history is divided into four major time periods: the Paleo-
Indian, from the time of documented entry into this area about
10,000 years ago until about 8,500 years ago; and the Archaic,
which is divided into Early, Middle, and Late phases.

The earliest periods are only slightly known from
archeological evidence. At Stratton Creek, a fragment of
spearpoint may date from the Paleo- Indian period. Both Marial and
Stratton Creek provide artifacts from the Early Archaic period,
from about 8,500 to about 7,000 years ago. These scant
archaeological traces indicate a way of life that was more mobile
than later periods, with small groups of people moving themselves
among various resources during the year.

vi

The subsequent Middle Archaic period lasted some 5,000
years, until about 2,000 years ago. The sites of Marial and
Stratton Creek document both the cultural stability of this
period, as well as the increasing pace of change as the era
closed. Throughout this era, both sites continued to function as
seasonal base camps for hunting and processing game and for
making and repairing stone tools, and to represent a mobile,
foraging way of life.

Yet cultural change is also evident during the Middle
Archaic. It may have been incremental at times, but by the end of
this era the stage had been set for a rapid expansion into the
more sedentary village way-of-life, with well defined cultural
territories and intense utilization of resources. New
technologies for processing and storing food were being refined
during the Middle Archaic, and populations were increasing.
Larger populations led to increased group cooperation and a more
complex social structure; together with developing technologies,
these changes laid the foundations for a new economic order.

Archaeologically, the beginning of the Late Archaic era in
southwestern Oregon is marked by the advent of the bow and arrow
approximately 2,000 years ago. The transition from the mobile,
forager subsistence pattern to the more sedentary collector
pattern intensified during the Late Archaic. Other cultural
hallmarks of this era include expanded trade networks with both
neighboring and distant groups, and societies in which social
distinctions were increasingly marked by wealth.

Archaeological excavations, such as those at the Marthaller
and Ritsch sites, stand out as clear examples of the Late Archaic
way of life. There is a marked increase in the number and variety
of tools, indicating more intensive resource use and longer
periods of occupation at a given location. Fishing appears to be
increasingly important, and sites reveal considerable heavy
"furniture", such as milling and grinding stones, as well as
substantial architecture. Clearly, these sites of the Late
Archaic had become villages, in the sense that people spent
substantial amounts of time at these locales and returned to them
year after year.

The reasons for the change into the more sedentary pattern
during the Late Archaic are not clear. Perhaps the best
explanation to fit the data is the interaction of a variety
factors, including: the changing environmental conditions of the
Middle Archaic that produced a more productive and predictable
food supply in the river valleys by the beginning of the Late
Archaic; a growing population that required an intensive
utilization of resources within more restricted territories; the
evolution of food processing and storage techniques, and of land
management practices (e.g., field and forest burning); and

Vll

expanded contacts through trade with societies outside of the
area .

The ethnographic way-of-life, documented for the period of
contact with Euro-Americans, may have differed from that of
earlier portions of the Late Archaic for several reasons. Prior
to the full-scale invasion of the Indian's ancestral territory
beginning in 1850, indirect Euro-American influences may have
profoundly changed native economic and social patterns. For
example, the effects of European diseases on Native American
cultures have been well documented. Although no direct evidence
exists for such epidemics in southwestern Oregon, they may well
have occurred early on and significantly altered the structure of
the native societies.

There are a number of sites along the river within the
project boundaries that date to the Late Archaic. Of that
number, a few can be directly associated with the ethnographic
era. The sites of Yawa-kh. Sumulkh, and Talda' cdan all have
direct, specific ethnographic references. Less well documented
evidence, but none the less suggestive, exists for the
association of the village of Da-ku-tee with the Marthaller site,
or possibly the Ritsch site; and the village of Hat ' onkh with the
site recorded near Jump-Off Joe Creek.

Early conflicts between the Native Americans and the
occasional fur trappers and traders who ventured through the
region in the 1820 's and 1830 's set the tone for future relations
between the two cultures. By 1850, settlers were beginning to
enter the Rogue River region in increasing numbers to take up
land. With the discovery of gold on Jackson Creek in late 1851,
the trickle of immigrants turned into a torrent. Within the
space of a few months, miners had stake claims throughout
southwestern Oregon and boom towns had sprung up along the rivers
and creeks. Increasingly, the Indians found themselves
dispossessed from their ancestral lands.

Conflicts between the cultures escalated until warfare broke
out in the Rogue Valley during the summer of 1853. In the
following spring of 1856 the final engagements of the war were
fought. Following surrender, those Indians who had survived war,
privation, and disease were forced to face another trial, removal
to a reservation. The early years at these new reservations were
a continuation of the degradations and deprivations that had
ravaged these traditional societies for the preceding decade.
Here they were forced to live in an unfamiliar environment among
people with different backgrounds and traditions, and suffered
many privations.

The twentieth century has been a long struggle from the edge
of extinction to the promise of a new era for descendants of the
original inhabitants of the Rogue River. The members of the

viii

Siletz and Grande Ronde fought back from the official termination
of their tribal status in the 1950 's to reclaim a portion of
lands allotted to them in the nineteenth century. Today, tribal
governments increasingly control their political and economic
future, and they have developed programs to revive cultural
traditions and to protect sites and landscapes of sacred value.

PART II: The Historic Era

The Rogue River's long story is about people who have sought
its wealth, isolation and beauty for at least 8,000 years. The
modern chapter of the stream' s history begins early in the
nineteenth century when Euro-Americans arrived in southwest
Oregon for the first time. Chapter six reviews the section of
river which extends between the Applegate River and Grave Creek,
and Chapter Seven traces the history of the rugged stretch of
river between Grave Creek and Mule Creek. Chapter Eight surveys
the entire study area since World War II.

Chapter VI: Applegate River to Grave Creek: Peter Skene Ogden,
Hudson's Bay Company chief trader, led an expedition of trappers
into the Rogue River Valley in the spring of 1827. The first
known explorer to journey through the region, Ogden' s party was
one of several trapping and exploring groups to traverse the area
prior to a local mid-century gold discovery. Others included the
U.S. Navy Exploring Expedition of 1841 and James Clyman in 1845.
In 1846 the Applegate expedition passed through Southern Oregon
while laying out the route of an emigrant road. These newcomers'
activities directly conflicted with the subsistence living
patterns of native peoples and within a few decades the ancient
culture disintegrated as seasonal rhythms were broken forever.

The discovery of gold in California in 1848 brought
Willamette Valley farmers- turned-miners south over the
California-Oregon Trail into the Rogue country. Prospectors
explored the Rogue River and its tributaries on their way south
and following the 1851 gold discovery near Jacksonville miners
swarmed over streams throughout the region. The abrupt influx of
miners into the Rogue country devastated local Indian bands.

Indian Mary Park in Josephine County marks the location of a
late 19th century inequity --a sad injustice to people who had
already forfeited everything. The land comprising the park once
belonged to "Umpqua Joe" who operated a ferry near the site. He
was allowed to remain on his land following removal of other
Indian peoples as a reward for warning Galice miners of an
impending Indian attack. After Joe's death in 1886, his
daughter, Mary, who had been born on the land about 1856,
received a grant with homestead rights from the federal
government. Following her death, however, local authorities
determined that she had no title to the land and her heirs were

ix

not recognized. This action constituted the last local instance
in which an Indian land-holding passed to Euro-American hands.

After Indians were removed from the Rogue country in 1856,
the region was fully opened to miners and farmers. Mining
developed at a steady pace along the Rogue as important gold
discoveries were made. Rich sites were found in bench gravels in
the vicinity of Pickett Creek, Hellgate Canyon, and Galice and by
the end of the nineteenth century extensive hydraulic mines
operated at several locations along the river.

The Flanagan Mine, located on the west bank of the Rogue
River about one mile upriver from Robertson Bridge, exemplifies
large hydraulic mining operations. Consisting of about 200 acres,
the Flanagan Mine site includes several mining faces, a large
area of tailings and the partial remains of water ditches.

Another long-active mine site, the Robert Dean Placer Mine,

was established on September 8, 1900 by Lydia Dean and H.A.
Corliss on the right bank of the river near Rand. By 1910 the
claim included one and three-quarters miles of ditch, a
reservoir, hydraulic giant and pipe, two cabins and six placer
excavations. The mine was run by Dean family members until 1940,

The Almeda Mine, just below the mouth of Grave Creek, was
one of the most extensively developed mines in southwestern
Oregon, with large quantities of copper, gold, silver, lead and
zinc extracted from nearly one and one half miles of underground
workings. During its principal period of operation between 1905
and 1916, production at the Almeda Mine was valued at more than a
hundred thousand dollars. Financial difficulties forced closure
of the mine in 1917.

On the heels of the miners came the farmers. The Donation
Land Claim Act, passed by Congress in 1850, offered settlers
free, farmable land. In its amended form, the act promised 320
acres of land to a married couple and 160 acres to a single
person if they would reside on the property and construct
improvements within a proscribed time. Several families chose
land along the Rogue River between the mouth of the Applegate
River and Pickett Creek and near the present site of Robertson
Bridge .

By 1875, farms were well established at Taylor Creek, at the
mouth of Jump-off Joe Creek, at Shan Creek, Limpy Creek, and
Dutcher Creek. After 1883 and the railroad's arrival at Grants
Pass, permanent settlement increased steadily. By 1900 houses,
barns, fences and roads characterized the gently rolling land
along the Rogue from the Applegate River to Pickett Creek.

Several floods affected residents along the Rogue during the
last half of the nineteenth century. During the winters of 1861

and 1890 particularly severe floods inundated fields along the
river plain west of present Grants Pass and destroyed
improvements, crops and roads from the Applegate to Jumpoff Joe
Creek.

By the early 2 0th century a rapidly improving road system
and increased auto travel combined to make the canyon more
accessible to tourists. Soon the beauty of the Rogue Canyon was
broadly recognized as visitors explored the back country. Zane
Grey, writer-adventurer who publicized the river's wild beauty,
brought wide recognition to the river with the publication in
1924 of Rogue River Feud, and in 1928 with his personal account.
Tales of Fresh Water Fishing.

During the 1920 's and 1930' s fishing lodges appeared in the
Rogue Canyon and adjacent areas. Some of the best quarters could
be found at "Speed's Place on the Rogue." This attractive, rustic
lodge was developed in 1928 near Galice at the former Galice
Consolidated Mine headquarters and remained open for several
years .

The Depression brought difficult times to the Rogue River
Canyon. Residents mined, cut wood and grew their own food to
survive. In 1933 the Civilian Conservation Corps (CCC) opened a
camp near the forest guard station at Rand in 1933. The Corps,
organized to provide work for citizens needing economic
assistance, substantially improved transportation and
communication. Under supervision of the U.S. Forest Service, the
crews built miles of truck trails, fought fire, and built bridges
within the remote reaches of the river canyon. Most CCC men left
the area by 1941 to serve in World War II.

Chapter VII: Grave Creek to Mule Creek: Along the stretch of
river between Grave Creek and Mule Creek, miners freely roamed
the Rogue Canyon after the removal of native peoples in 1856. Men
and equipment were scattered along Grave Creek, Whisky Creek, and
China Gulch. Only a few men, however, lived permanently in the
Canyon during the 1860 's and early 1870' s. The few small
terraces above the river were just wide enough to hold a shelter
and equipment and most of the miners moved frequently. Chinese
miners worked many locations in the Rogue Canyon after 1855 and
continued to work placer deposits on river terraces long after
Euro-Americans had extracted the more easily mined gold.

About 1880, placer miners who had settled in for a season
about three miles downriver from Grave Creek, planned a sturdy,
permanent dwelling. They selected a site for their cabin on
Whisky Creek about 1000 feet above the river and began
construction of a twenty-by- twenty foot residence. This dwelling,
known now as the Whisky Creek Cabin, housed a succession of
miners for almost seventy-five years. While flood, fire and

XI

demolition have eliminated most dwellings in the Rogue River
Canyon, the Whisky Creek Cabin remains a distinct, rare resource.

Large-scale hydraulic mining increased near Mule Creek when
the Red River Gold Mining and Milling Company, an Indiana based
corporation, purchased several claims at the mouth of Mule Creek
in 1906. The company operated for six years before closing in
1912. The impact of early twentieth century hydraulic operations
is still evident in the Rogue corridor in huge piles of mine
tailings, ditch remainders, and scattered lumber left from
improvements and wooden flumes.

Hardrock or lode mining, which had begun in both the Gal ice
district and the Mule Creek district before 1900, developed
rapidly after 1900. Lode mining, like hydraulic mining, required
substantial capital and frequently demanded the backing of
corporate groups. Processing ore from lode mines required that it
be drilled or blasted from the vein, loaded into cars and taken
to the surface.

Although placer and lode mining continued through the
1920' s, the Depression brought an influx of people into the
Canyon. As other sources of income declined these newcomers were
enticed by an increase in the gold price in 1934 and the chance
to carve out a subsistence living in the Rogue Canyon. Men such
as J.C. ("Red") Keller and Lou Martin spent years working river
crevices and bars .

World War II brought an end to most of the subsistence
living created by the depression era miners. The sale of gold was
halted in 1942 and military service or job opportunities in war-
related industries took many of the younger men away. Most never
returned to the Rogue to live. In the years following the war
tourism increased as anglers, hikers and boaters flocked to the
river.

People who settled on the few accessible places in the Rogue
River Canyon were isolated from the rest of the world, but not
from each other. They worked long, hard, bone-wearying hours and
most of their energy was spent acquiring food, shelter, and
clothing. These people survived with very little money in a
"subsistence" economy, in which they grew, hunted, fashioned or
bartered for the bare necessities of life.

The gently sloping land near the mouth of Mule Creek had
supported people intermittently for thousands of years. Here the
Billings family established a ranch and a community where,
isolated by the rugged terrain and unnavigable river, they and
other residents exploited the* resources around them, including
fish, game, fuel and gold. These families raised their own stock
and food, crafted most of their own tools, and traded or bartered
for what they could not manufacture. In 1883 the Oregon and

xli

California Railroad construction reached a point on the West Fork
of Cow Creek, offering a closer and more convenient supply point
for residents along the Rogue River and through the surrounding
hills.

George Billings, whose mother, Adeline, was a Karuk Indian,
was eligible for a homestead allotment, and in 1895 settled
permanently at Mule Creek. About 1900 he built a larger home
using lumber milled from trees cut from his land. Completed
about 1903, the residence stands on what is now known as the
Rogue River Ranch. In 1908 the Billings family built a large barn
to house pack mules and equipment. Parties, elections and church
services were held for the family and neighbors on the upper
floor of the barn.

In 1931 George Billings sold the ranch to Stanley Anderson,
a Hollywood businessman. The new owner enlarged the house, added
a caretaker's residence, bunkhouse, tackhouse, woodshed, storage
shed, and restored a blacksmith shop. In 1970 the entire two-
hundred acre parcel was sold to the U.S. Government under the
National Wild and Scenic Rivers program.

Chapter VIII: Postwar to Recent Years (1945-1995): After World
War II, the Rogue between the Applegate River and Mule Creek
changed in important ways. Upriver, large farms were divided into
smaller parcels and road construction made transportation easier
within the settled areas. In the Rogue Canyon, Civilian
Conservation Corps roads allowed new access to previously
isolated areas and trucks replaced the pack trains. Galice
gradually developed into a supply point for fishing, boating, and
rafting groups. At Marial, older members of the large Billings
clan died in the years following the war. The post office at
Marial closed in 1954. Now anglers, boaters and hikers visited
the canyon each season and lodges flourished at Black Bar and
Marial .

As the tourist business increased, interest rose in making
river passage easier. Although excellent boatmen had maneuvered
the difficult rapids and tumultuous water through the late
nineteenth and early twentieth centuries, several rocky locations
formed difficult barriers. Following World War II, Glen
Wooldridge combined his skill at boatbuilding and the use of
dynamite to open the Rogue to traffic by increasing the blasting
activities which had extended over decades. Gradually the river
was fully opened to commercial fishing and raft trips.

Floods continued to occasionally cause serious damage along
the Rogue. Flows were high enough during the major flood years
of 1955, 1964 and 1974 to destroy bridges, roads, built
improvements and to inundate agricultural lands and stream
courses. The completion of dams at Lost Creek on the upper Rogue
and on the upper Applegate River, have greatly reduced the

xiii

chances of major floods on the middle and lower portions of the
Rogue. While protecting soil and improvements, the dams have also
dramatically altered the river's ecological balance.

Following World War II, the U.S. government took a new role
in overseeing human uses of the Rogue River Canyon. A federal
agency had been present in the canyon since the creation of the
U.S. Forest Service in 1906. By 1920 there were two Forest
Service Guard stations within the river canyon, one at Whisky
Creek and one between Meadow Creek and Horseshoe Bend. Fire
detection and trail construction were major tasks assigned to the
Forest Service.

The Bureau of Land Management, established in 1946 out of
the Department of the Interior, took over administration of non-
National Forest timber lands within the watershed in the late
1940 's and assumed responsibility for managing the river
corridor. The Wild and Scenic River Act was passed in 1968 and
the Rogue was subsequently listed. In order to restore the canyon
to natural conditions the agency began identifying mining claims
and cabins which were illegally held. Most of these dwellings
stood on mining claims and were used as vacation homes. Cited
under "illegal trespass," title to these lands was purchased or
claimed from owners. In 1995 few, if any, of these structures
were left along the river.

The Wild and Scenic River designation of the river was
assigned between the mouth of the Applegate and Marial, on Mule
Creek in Curry County and this portion of the river corridor was
placed under the jurisdiction of the Bureau of Land Management.
The segment was divided into two designations for management
purposes; Recreational and Wild. The Recreational section of the
Rogue extends from the mouth of the Applegate River to Grave
Creek. The Wild section begins at Grave Creek and continues to
Watson Creek, a distance of thirty-two miles. (Bureau of Land
Management responsibility ends at Mule Creek.)

The Rogue's national reputation as an excellent salmon and
steelhead fishing stream, as well as a river with outstanding
white water boating and rafting opportunities, brings thousands
of visitors each year. Fishing, float trips, camping, hiking,
picnicking, swimming and sightseeing are all enjoyed within an
area of great natural beauty.

The Rogue River's story has evolved over thousands of years.
For centuries prior to 1820 native peoples adapted successfully
to their environment. With Euro-American occupation came the
destruction of the sites used by Indians for subsistence
activities. The qualities that drew people to the area centuries
ago continued to attract. Although the modern period reflects
only a moment in history, the ways in which we use and care for

XIV

the Rogue will determine what remains for future generations who
are drawn to its banks for sustenance or seclusion.

XV

PART I: THE INDIANS AND THE RIVER

I . INTRODUCTION

The Medford District of the Bureau of Land Management (BLM)
contracted with Cascade Research of Ashland, Oregon, in September
1994 to summarize the history and prehistory of the BLM managed
Wild and Recreational section of the middle course of the Rogue
River. This overview is the third and concluding phase of a
study of cultural resources undertaken by the BLM and Cascade
Research. Previous studies have focused on a review of documents
and literature concerning the history and archaeology of the
study area, and the field recording of a sample of sites that
exist along the river.

The purpose of this overview is to synthesize the
information generated in the two previous studies and to place
the documented cultural resources along the river into an
historical context that links human activity with the natural
environment of the Rogue River over the past 10,000 years. This
study is directed to the broad themes of economic, social, and
technological change. When feasible and appropriate the focus
has been narrowed to individuals, specific sites, or isolated
events, which serve to illustrate a particular way of life or an
era of time. It is the intent of the authors to provide the
managers of public lands with a document that will be of use for
future planning, as well as to inform those who have an interest
in history of the cultures and people who have called the Rogue
River their home.

The study area specified in the contract, 46.5 miles along
the Rogue River in southwestern Oregon, extends from the
confluence of the Rogue and Applegate rivers (approximately five
miles west of the city of Grants Pass in Josephine County) to the
settlement of Marial at Mule Creek in Curry County 48.5 miles
inland from the Pacific coast (Figure 1) . The boundaries of the
Wild and Scenic River designation (generally one-quarter mile on
either side of the river) also bound the current study area.
This portion of the Rogue River is divided into two designations
for management purposes: Recreational and Wild. The Recreational
section of the Rogue River is the portion that is readily
accessible by road or railroad and may have some development
along the shoreline (Purdom 1977) . The Recreational section of
the river managed by the BLM (termed Hellgate Recreation Area)
extends from the mouth of the Applegate River to Grave Creek (27
miles) . The Wild section of the Rogue (defined as free of
impoundments and generally accessible only by trail, with
shorelines essentially primitive and water unpolluted [Purdom
1977] ) begins at Grave Creek and continues to Watson Creek, a
distance of 32 miles. However, the Bureau of Land Management's
jurisdiction of the Wild section of the Rogue River only extends
to Mule Creek, approximately 19.5 miles downriver from Grave
Creek.

to Portland

LEGEND

..... Wild River
ooooo Scenic River
Recreational River

Figure 1 Project Location Map (After Purdom 1977)

2

II. ENVIRONMENT

Introduction

Environmental knowledge is crucial to understanding the
forces that determined where people lived and how they made their
living, as well as the economic developments that have affected
the land along the Rogue River. The availability of mineral
resources, the variety of plants, the abundance of wildlife, the
constraints of the terrain and the climate, and the power of the
river itself all influenced the location and duration of human
occupation along the river's shore.

The Rogue River begins its journey to the Pacific Ocean at
Boundary Springs within Crater Lake National Park on the western
slope of the High Cascades at an elevation of approximately 5,260
feet above mean sea level. From its headwaters, the Rogue flows
generally southwest to the broad Rogue River Valley near the City
of Medford at an elevation of about 1,300 feet. At this point
the river gradient flattens and the river only drops 500 feet in
elevation over the next 40 miles to the town of Galice. Five
miles west of Grants Pass, the Applegate River joins the Rogue
and it shortly turns to the northwest toward Galice,
approximately 20 miles downriver. Six miles north and downriver
of Galice the river is joined by Grave Creek where the Wild
section of the river begins. From this point (approximately 700
feet above sea level) the river makes a sharp bend to the west;
first to the northwest and then to the southwest before emptying
into the Pacific Ocean (Purdom 1977) .

Topography and Vegetation

The section of the Rogue River managed by the Bureau of Land
Management encompasses a variety of geologic formations,
vegetation regimes, and geographic landforms that add to the
complexity of human settlement patterns within the boundaries of
the project area. From the mouth of the Applegate River, at the
upriver boundary of the project area, to the Hellgate Canyon
area, the valley formed by the river is generally more broad and
open than further downriver, although there is rugged terrain and
narrow canyons in this section of the river as well . The broad
river terraces along this upper, drier stretch of the Rogue River
support a vegetation regime dominated by oak woodlands. Pacific
madrone, and manzanita. Mixed evergreen and conifer overstory
{e.g., big-leaf maple. Pacific madrone, black cottonwood,
ponderosa pine, and Douglas-fir) are interspersed with the oak
woodlands (white oak, black oak, and canyon live oak) on the
north facing slopes of the more mountainous sections along the
river; dense riparian vegetation is common along the shoreline.
Below Galice, on the Wild Section of the river within the Rogue
River Canyon, the terrain is steeper with only occasional large
terraces. Although the forest species are similar, aspect is a

more significant factor along this part of the river; the north
facing slopes are dominated by a mixed conifer forest, where the
drier, southern aspect terrain supports more drought tolerant
overstory species (e.g., oak and Pacific madrone) (Purdom 1977) .

Geology

Gold-bearing geologic formations are present within much of
the project area. The remains of extensive hydraulic placer
mines in the Recreational Section of the river, as well as lode
mines in the Galice-Almeda area attest to the fact. Only in the
Wild Section of the river between Horseshoe Bend and Marial,
within the Dothan Geologic Formation, are gold deposits absent
(Purdom 19 77) . The only gold found in this area comes from gold-
bearing gravel washed down from lode deposits further upriver.
The geology of the area is complex and has influenced the course
of the river and its gradient, as well as areas of human
settlement. (For a detailed description of the geologic history
of the Rogue River, see Purdom 1977) .

Fauna

The Rogue River is home to a variety of wildlife. Black-
tail deer, black bear (also formerly grizzly bear) , elk, cougar,
as well as smaller mammals such as squirrels, chipmunks,
raccoons, weasels, skunks, and mice inhabit the forest and
valleys along the river. A number of amphibians and reptiles,
including salamanders, newts, frogs, lizards, and pond turtles
dwell along the river's edge. Significant bird species found
within the project area include ospreys, blue herons, ducks, bald
eagles, woodpeckers, grouse, and quail (Purdom 1977) .

Outside of the gold-bearing formations found along the Rogue
River, the presence of anadromous fish (steelhead trout and
salmon) have had the greatest effect on the human use and
occupation of the Rogue River region. From a primary, dependable
source of protein for prehistoric inhabitants to a tourist
industry in the twentieth century, salmon have supported and
encouraged the settlement and use of the river by a wide variety
of people.

How far back in time Native American people utilized this
riverine resource is an unsolved question, although the numerous
village sites located on bars and terraces near riffles and falls
along the river suggests a long history of dependance on this
food source. The history of the modern recreational boom on the
Rogue River can also be attributed to the once prolific
anadromous fish runs. Floating the Rogue River originated (as
did the construction of lodges along the river) in response to
the demand of anglers. Indian village and fishing sites,
recreational summer homes and commercial lodges, and transient

4

use of the river all have been predicated upon the fisheries of
the Rogue River.

Climate

The climate of southwestern Oregon has not been static.
During the Holocene (the past 10,000 years) shifts in temperature
and precipitation affected the type and extent of vegetation, the
viability of stream and river flows, fish and animal populations,
and human access to higher elevations. Although direct evidence
of the past climate and environment is lacking for southwestern
Oregon, the broad patterns of climate change experienced
throughout the American West can serve as a model. In general,
at the beginning of the Holocene, temperatures were rising and
the climate was warmer and drier than today. This trend
continued until sometime after 6,000 years ago, when wetter and
cooler conditions began to appear. During the past few thousand
years modern climate patterns and vegetation regimes have
prevailed. However, during this period the environmental forces
have not been constant. Fluctuating cycles of drier or wetter
conditions, varying in duration, characterize the modern climate
pattern.

The effects of a prolonged drought over thousands of years
during the early part of the Holocene in southwestern Oregon
would have created differences in the environment from what we
know today. Since early inhabitants were closely linked to the
existing environment, these differences affected their way-of-
life. Warmer temperatures and less precipitation than today,
for example, would have promoted the spread of the oak/chaparral
grasslands of the interior valleys towards higher elevations, and
may have affected the quantity and predictability of a staple
food, acorns. Other Native American staple food supplies (e.g.,
camas) may have been restricted to higher, more moist elevations
(Hannon 1992) . Lower snowpack levels and resulting runoff could
have affected both the number of seasonal anadromous fish runs
(salmon and steelhead trout), as well as their distribution.
Spencer (1991) postulates that lower stream flows in the spring
and summer would have made steep stream gradients and low falls
effective barriers to migrating fish, restricting there inland
range. Warmer stream water temperatures would also have limited
fish populations. Thus, the only dependable major fish run may
have occurred only during the winter, and might not have reached
as far inland as historic fish migrations.

This long period of drier and warmer conditions in
southwestern Oregon began to change at some point in the mid
Holocene. The onset of wetter, cooler conditions gradually
changed vegetation patterns, as well as the quantity and
distribution of game animals and migrating fish. For example,
higher year-round stream flows would have benefitted migrating
fish populations, and the harsher winters would have driven many

game animals to the lower valleys. The longer winters and
heavier snowpack restricted use of the higher elevations by
native peoples to a shorter part of the year. Increased
precipitation made for a more abundant and predictable acorn
crop, as well as allowing camas to spread to moist areas in lower
elevations. These changes from dry to moist conditions in
southwestern Oregon may have made the low elevation, interior
valleys more attractive for human settlement; more resources were
now available on a predictable basis. These environmental
conditions became the basis for increased human use of the study
area .

The present-day climate along the Rogue River, although
subject to cyclical fluctuations, can be generally characterized
as having mild, wet winters and hot dry summers, making year-
around settlement possible throughout most of the project area.
Rainfall is greatest along that portion of the river closest to
the coast. Rainfall at Illahe (about 18 miles downriver from
Marial) averages 80 inches of rain per year, while at Grants Pass
only 26 inches are received annually (Purdom 1977) . Most of the
precipitation falls in the winter months. Light snowfall is not
uncommon within the project area, and heavy snows occasionally
occur. Considerable fluctuation in annual temperatures and
precipitation is predictable.

Floods

Periodic flooding within the Rogue River basin has had
devastating consequences on the cultural environment. The rare
combination of a warm southwesterly storm system with several
inches of rain and an existing snowpack has, at times, produced a
massive melt and runoff causing major floods along the Rogue
River and its principal tributaries (Highsmith 1979) . High water
has occurred frequently on the Rogue through the years, and
indications are that floods similar to modern ones occurred
historically.

Accumulated snow in December, 1852, subsequent warming
weather and heavy rains in January, 1853 combined to cause severe
flooding in the Bear Creek Valley and along the Rogue River in
what is now Josephine county. In March, 1859 the Rogue River and
other southwestern Oregon streams swelled following severe rains
and "caused a flood in almost all streams" (Crescent City Herald
1858) . The flood of December 1861 was the largest flood of
record on the Rogue River. In that year severe flooding
inundated fields along the Rogue River plain west of Grants Pass
and destroyed improvements and crops along the Rogue River in the
agricultural section from the Applegate River to the mouth of
Jump-Off Joe Creek. Major floods of record also occurred in
1890, 1927, 1955, 1964, and 1974. Less severe flooding- -though
still causing major damage--took place in 1864, 1881, 1893, and
1903 .

6

River flows were high enough during these major flood years
to destroy bridges, roads, built improvements, mining structures,
and to inundate agricultural lands and stream courses. No
written record exists of flood impact on human improvements,
soil, vegetation, or aquatic life before Euro-American settlement
and development, although certainly catastrophic one-hundred year
floods occurred then, as in the recent past. These periodic
floods either deposited layers of silt and sand over the sites of
Native American encampments preserving them for future
archaeological investigation, or alternatively, eroded the cut
banks and terraces along the river, sweeping the cultural remains
of centuries downriver to the sea.

The construction of dams in the 1970' s at Lost Creek on the
upper Rogue River and on the upper Applegate River have greatly
reduced the chances of major flooding episodes along the middle
and lower portions of the Rogue River, although not without a
cost to the ecological balance of the river. Periodic flooding
helps to cleanse the gravel that serves as spawning grounds for
anadromous fish, as well as thinning out the riparian vegetation
along the river shore (Purdom 1977) . From a culture history
point-of -view, the lessening of flood action may well serve to
protect fragile archaeological sites from further erosional
damage or from burial under sands and silts deposited in areas of
slack water.

The forms, forces, and fluctuations of nature have
influenced human occupation along the shores of the Rogue River
for thousands of years. From the broad open valleys on the upper
reach of the project area, to the narrow river canyon below Grave
Creek, the river, the underlying geologic formations, the
topography and natural vegetation, and perhaps most important,
the fisheries of the Rogue River, have affected were humans have
lived (settlement patterns) and the means by which they sustained
themselves (subsistence patterns) . From the mouth of the
Applegate River to Grave Creek the broad terraces located by
falls and riffles near good fishing, with access to staple
vegetable crops in the valleys and foothills of the Rogue River
Valley, provided the most favorable settlement habitat for the
Native American way-of-life. Only the occasional riverside
terrace further downriver in the canyon allowed them sufficient
space and access to resources to maintain encampments.

For the Euro-American settlers, the broad terraces of the
upper reach of the project area below the Applegate River
provided the fertile soils and access to markets that were
necessary for an agricultural economy. These same locales today
continue to sustain agricultural activity and, increasingly,
residential development. Miners were less in need of open space.
They settled where they could near the source of their
subsistence, gold. Small benches, rocky bars, or high terraces
provided the necessary settlement prerequisites for the itinerant

miner, as well as for the later corporate enterprises. In recent
decades the secluded cabins and small settlements of gold miners
have yielded to supply centers, fishing lodges, and campgrounds
of the modern recreational era. The force and power of the river
has been altered in modern times. Dams no longer allow periodic
floods to devastate the works of a society or to restore the
river's natural balance. Although our industrial society has
partially tamed the force of the river, the environment will
continue to influence human occupation and use of the area in the
future, as it has for past ten thousand years.

III. ARCHAEOLOGICAL METHODS, THEORIES, AND ISSUES

Introduction

The portion of the Rogue River within the current study area
is centered within a larger geographical area stretching from the
Pacific Ocean on the west to the High Cascades in the east, and
from the Klamath River in northern California to the Umpqua River
basin situated to the north. Throughout history, the native
people of this region have shared a similar environment and
culture history. Prior to examining the specifics of the history
of the project area a brief review of archaeological methods, and
the current theories and issues regarding the lifeways of the
native inhabitants of the region and how they may have changed
over time, will provide the necessary context to evaluate the
archaeological remains and ethnographic record along the Rogue
River.

Archaeological Methods

What little is known of the lifeways and economic systems of
the people of southwestern Oregon during the 10,000 years of the
Archaic period is derived from archaeological investigations,
climate and environmental models, and extrapolations from
ethnographic and historic records. All of these sources suffer
from a lack data. Only in the past few decades has sufficient
archaeological information been generated to even begin to assess
the culture changes that have occurred since the beginning of
human occupation of the region. The further removed in time
from the present, the more difficult it is to document and
characterize the way-of-life. However, in recent years the pace
of archaeological research in southwestern Oregon has increased
dramatically, to the point where the broad outlines of the past
have emerged from the shadows.

One of the first tasks of archaeologists in any region is to
develop a chronology. Archaeologists must sort through the
remains of vanished peoples and make sense of the stylistic and
functional changes in their tools over time. Only when
archaeological sites within a region have been firmly dated can
the more complex process of reconstructing a picture of past
lifeways be undertaken. Developing a cultural chronology for
southwestern Oregon has been a primary goal of many investigators
over the course of the past sixty years.

Several techniques are available for establishing the dates
of archaeological sites along the river, and identifying the
artifacts associated with specific time periods. These include
stratigraphy, radiocarbon dating, cross-comparison of
stylistically-dated artifacts, and obsidian hydration analysis.
These dating methods are briefly described below.

Stratigraphy: This is the relatively simple principal that
older artifacts will be recovered from an archaeological
site at a greater depth (or level) than younger artifacts.
Much of the energy that archaeologists spend in excavating a
site is based on this technique. Careful excavation of
archaeological sites can often determine the original
relative position of artifacts, and hence, the chronological
relationship among them. Once stratigraphic layers are
determined, the relative age of artifacts associated with
them is apparent. In southwestern Oregon, however,
stratified sites are the exception rather than the rule.
There are many natural agents, such as borrowing rodents,
that change or modify the cultural deposits in the ground.
Determining the absolute dates of artifacts, or groups of
artifacts, requires one of the other dating techniques
described below.

Radiocarbon dating: Radiocarbon dating measures the rate of
decay of an isotope of carbon (C-14) that is present in all
living organisms. When a plant or animal dies, the C-14
present in the organism begins to decay. This rate of decay
can be measured, thus dating the death of the organism in
absolute terms. Bone, shell, and plant materials (e.g.,
charcoal from a fire hearth) that are directly connected
with archaeological artifacts or deposits are collected
during the excavation of a site. Analyzed C-14 samples are
used to date the stratigraphic levels within a site, as well
as the artifacts associated with those levels. Radiocarbon
dating is the most reliable technique archaeologists have
for dating artifacts and sites. In southwestern Oregon,
however, the seasonal cycle of winter moisture and summer
heat promotes the rapid decay of most organic materials. As
a result, many sites do not have materials suitable for
radiocarbon dating.

Artifact cross-comparison: This is a dating technique that
builds on the absolute dates generated by radiocarbon
analysis. Once a distinctive style of artifact (e.g., a
particular projectile point-"- style) , or groups of artifacts
(known as an artifact assemblage) , have been dated at a
number of sites within a region, a similar artifact or
artifact assemblage discovered in a context devoid of either
stratigraphic relationships or radiocarbon dating potential
can be dated to the same era based on stylistic

-'• Projectile point is the generic term for chipped-stone
artifacts used to tip an arrow, dart, or spear. Archaeologists
distinguish arrow points from dart or spear points by the overall
size and weight of the artifact, as well as by the width of the
hafting element (i.e., the portion of the point that was attached
to the shaft) .

10

similarities. A good example is the Clovis spear point of
the Paleo- Indian era. A number of these distinctive
projectile points have been recovered in the southwestern
United States associated with radiocarbon dates of between
10,500 and 11,500 BP . Thus, the isolated finds of
stylistically similar Clovis points in Oregon are assumed to
date from the same time period.

The use of artifact cross-comparisons has played a
major role in the development of the chronology of
southwestern Oregon due to the lack of well-stratified sites
and the paucity of radiocarbon dates. However, as we shall
see, this heavy reliance on the artifact cross-comparison
dating technique within the region may have obscured the
complexity of the archaeological record in southwestern
Oregon.

Obsidian Hydration: When obsidian^ is broken or flaked,
which happens during the manufacture of stone tools, a fresh
surface of the stone is exposed and water then begins to be
absorbed into the exposed surface. This absorption of water

(hydration) creates a rim or band on the surface of obsidian
that can be detected under high-powered magnification. The
longer the surface has been exposed, the thicker the
hydration rim. In order to measure the thickness of the
hydration rim, a small section of an obsidian artifact is
cut, removed, and ground to a thickness of between 30 and 50
microns^ , and then the hydration rim is viewed and
measured under powerful magnification. There are factors

(not all of which are completely understood) that affect the
rate of hydration. The two principle factors affecting
hydration are the original source of the material and the
average annual temperature of the location where the
artifact has been deposited. Because different kinds of
obsidian hydrate at different rates, knowing the original
source location is critical for accurate comparison.
Chemical analysis of obsidian can accurately determine where
the material originated. Although hydration rates have been
used principally as a relative dating tool, recent work

(e.g., in the Elk Creek drainage on the upper Rogue River

[Nilsson and Kelly 1991] and in northeastern California's
Medicine Lake Highlands to the south [Cassidy 1992] ) has

^ Obsidian is a volcanic glass favored by Native Americans
for making chipped-stone tools due to its flaking qualities and the
resulting sharp edges produced. The nearest sources of obsidian
are located in eastern and central Oregon and northeastern
California .

^ A micron is a millionth part of a meter (39.37 inches) .

11

produced absolute chronologies that appear consistent with
other means of dating archaeological sites.

Prior to examining some of the current theories regarding
chronology and culture change in southwestern Oregon, one final
technique of value to archaeologists needs to be examined: the
ethnographic analogy. Ethnography is the study of the culture of
a community or a group. Beginning in the late nineteenth
century, anthropologists began efforts to reconstruct the
lifeways of the Indian peoples as they existed prior to the
arrival of Euro-Americans. Added to this systematic study of the
Indian cultures of the region are the historical observations,
diaries, and notes of early explorers and settlers who recorded
fragments of the native peoples' way-of-life. Both the
ethnographic records and the historical accounts relating to
Native Americans of the region have been synthesized in recent
years (Gray 1987 and LaLande 1990) . These ethnographic and
historical accounts provide archaeologists with a valuable tool;
a glimpse of the way-of-life prior to the intrusion of western
culture. A discussion of the ethnographic lifeways of the
Indians of the region is presented in Chapter IV.

Regional Cultural and Chronological Models

The archaeologically accepted date for human entry into the
Pacific Northwest is at the end of the last Ice Age between
12,000 and 14,000 years ago. A radiocarbon (C-14) date of 13,200
BP (Before Present) was obtained from cultural deposits at Fort
Rock Cave in central Oregon. By 11,500 BP small groups of people
had spread across the continent pursuing the herds of large game
that inhabited North America in the moist, post-glacial
environment. These highly nomadic, big-game hunters, known as
Paleo- Indians, used a distinctive style of spear point named for
the small town of Clovis, New Mexico near where they were first
discovered. These large, finely made Clovis spear points have
been found associated with large mammal species (megafauna) , such
as mammoth, horse, camel, and giant bison that once roamed the
continent. Isolated Clovis points (although not associated with
megafauna) have been reported in southern Oregon from near
Roseburg, Butte Falls, Hyatt Lake, and the Oregon-California
border, confirming the presence of Paleo- Indians in the region
during this early period.

With the warming of the climate in the post-glacial period
and the resulting change in vegetation, as well as the extinction
of large animal species, the nomadic, big-game hunting way of
life of the Paleo- Indians gradually shifted to a more broad-based
hunting and gathering existence that archaeologists term the
Archaic period. The Archaic way-of-life, which began around
10,500 BP, lasted (in the Pacific Northwest) until contact with
Euro-Americans in the nineteenth century. The economy of this
period was based on a wide variety of plant, animal, and fish

12

resources. New techniques for gathering, processing, and storing
foods were developed over the span of the Archaic era, as were
new technologies in weapon systems. Settlement patterns,
domestic architecture, social structures, and land management
practices reflected the changes in the natural environment and
the pressures of a steadily increasing population.

Within this generalized historical framework, the changes
that occurred in southwestern Oregon Indian societies are a
matter of on-going debate. The lack of data in the region
concerning the Paleo- Indian era allows little room for
disagreement. The generalized, nomadic hunting patterns
documented elsewhere across North America are assumed by most
researchers to have applied to southwestern Oregon as well. It
is for the long span of the Archaic era, from perhaps 10,500 BP
to the contact with Euro-Americans (ca. 1830-1850) that differing
models have been developed to explain the changes in population,
land use patterns, social organization, food resource
utilization, and technological developments. The following
discussion focuses on three different interpretations of the
archaeological record in southwestern Oregon during the Archaic
era .

Region-Wide Cultural Patterns: This model of the archaeological
record for southwestern Oregon and northern California was first
proposed by Thomas Connolly (1986) to explain what he saw as both
consistency and change in artifact assemblages across the region
for the past 9,000 years. Connolly examined artifacts from a
number of sites located from the Pacific coast to the foothills
of the Cascades, and from the Umpqua River in Oregon to the
Klamath River basin in California. Based on this analysis, he
proposes three different patterns in southwestern Oregon
representing two distinct ways of life.

The oldest and longest lasting of these patterns, deemed the
Glade Tradition, was defined on the basis of an artifact
assemblage composed of broad-necked projectile points (both side-
notched and stemmed) , leaf-shaped projectile points, and
contracting stem points. Other artifacts defining this tradition
include stone bowl mortars; hammer/anvil stones; thick, chipped-
stone endscrapers; and river cobbles with worn edges (Connolly
1990) . The projectile points are assumed to represent the use of
the thrusting spear and the dart and atlatl (the atlatl is a
wooden shaft used to propel a spear or dart) .

Connolly argues that these archaeological remains are
representative of a way of life that was characteristic of small
bands of people who were mobile hunters and gatherers (also
termed foragers) . These small groups were oriented to the
resources of land (e.g., hunting game and collecting plants),
rather than placing a heavy reliance on the region's fisheries.
Due to the fact that these small groups were more or less

13

constantly on the move in search of game and ripening plant
crops, they did not invest time or energy in either the
construction of substantial houses or villages, or in preserving
and storing food for later use. Rather, their settlements were
temporary and scattered over a territory large enough to sustain
their needs during the course of the changing seasons .

Based on the stylistic similarities of artifact assemblages,
radiocarbon dates, and obsidian hydration analyses, Connolly
postulates that the Glade Tradition may have persisted in remote
areas of the region until the Late Archaic era. However,
beginning about 1,700 BP another way of life, substantially
different from the preceding Glade Tradition, made its appearance
in the interior region of southwestern Oregon and northern
California. Connolly refers to this new lifeway as the Siskiyou
Pattern. The Siskiyou Pattern is similar to the way-of-life as
practiced by the region's Indian groups at the time of contact
with the first Euro-Americans. During the Siskiyou pattern era,
people resided in semi -permanent villages located along the major
river courses of the region. Their existence depended on
seasonal fish runs (salmon and steelhead) , as well as on hunting
and a more intensive utilization of a variety of plant resources.
Although mobile foraging parties still roamed the upland areas
during part of the year, food preservation and storage in winter
villages became the dominant economic pattern.

The artifact assemblage from sites representing the Siskiyou
Pattern reflects economic, social, and technological change.
Instead of large, broad-neck dart points, the smaller, lighter,
narrow-neck points used to tip arrows dominate; stone bowl
mortars are replaced by milling stones and hopper mortar bases;
the first use of ceramics are documented; stone net weights are
used for fishing; and people dwell part of the year in semi-
subterranean pithouses (implicit in this new pattern is an
increase in regional population). Connolly (1990) attributes
this rather dramatic change in lifestyle in part, to increased
contacts with people from east of the Cascades.

The Gunther Pattern, which occurred somewhat later on the
coast, paralleled the development of the Siskiyou Pattern in the
interior. Radiocarbon dates from sites along the southern Oregon
and northern California coast date the beginning of this pattern
to approximately 1,100 BP. In most respects, the Gunther and
Siskiyou Patterns are quite similar. The primary differences
relate to projectile point form (small, concave-base projectile

■* Pithouse is the term used by archaeologists to describe the
remains of the housing used by Native Americans of the region.
Winter dwellings of Indians in southwestern Oregon were plank or
bark covered structures built over shallow excavations. It is
these excavations or pits that identify where houses once stood.

14

points are prominent on the coast) , and to the use along the
coast of more artistic forms of ground-stone implements (pestles,
clubs, and mauls) . On the coast, as inland, an increase in
population is presumed and semi -permanent villages became the
norm. The geographic extent of the Gunther Pattern, as well as
the date of its inception, corresponds with the territory
inhabited by speakers of the Athapaskan and Algic languages who
are thought to have arrived in the region between 1,500 and 1,000
years ago (Aikens 1993).

Regional Cultural Chronology: Based on archaeological
investigations within the Elk Creek drainage on the upper Rogue
River, Pettigrew and Lebow (1987) (later refined by Nilsson and
Kelly [1991] ) have proposed a cultural chronology for
southwestern Oregon that is primarily based on stylistic changes
in projectile point forms over time. These researchers analyzed
the artifact collections from dated sites in the Rogue River and
Coquille River drainages of southwestern Oregon, as well as data
collected from a number of excavated sites along Elk Creek.

Their view of the culture history of the region divides the
Archaic era into four time phases, beginning after the Paleo-
Indian era ca . 10,500 BP (Figure 2). (The following summary
simplifies the chronological scheme proposed by Pettigrew and
Lebow, but retains their essential defining attributes) . The
earliest Archaic period is termed the Applegate Phase (10,500 to
8,500 BP) and is based on the artifacts recovered from only one
site, 35JA53, located in the Applegate River drainage. Artifacts
and characteristics of the site (and phase) , include large,
broad- stemmed projectile points, edge-worn river cobbles (thought
to have been used in hide working) , low obsidian use, and very
few end- scrapers (used for processing game and hides) . No
radiocarbon or obsidian hydration dates are associated with the
data. Dating of the site was based on geological information and
cross-comparison of projectile point styles with artifacts from
outside of the region. However, recent excavations along the
Umpqua River (O'Neill 1992) have uncovered similar projectile
points, as well as edge-worn cobbles, beneath a layer of Mt .
Mazama volcanic ash, which dates that site to before 6,800 BP.

Following the Applegate Phase is the Marial Phase, which is
proposed for the period between 8,500 to 4,500 BP. This era is
represented by large, leaf- shaped projectile points, a
significant number of end- scrapers, and a relatively high use of
obsidian for the manufacture of chipped-stone tools. During the
earlier part of the phase, edge-worn cobbles are prevalent; in
the latter half of the phase their frequency declines. The
overall size of the leaf-shaped points also decreases in the
later stages of the phase. A distinctive projectile point, the
McKee Uniface, is also associated with this time period. Dating
of this phase relies heavily on the radiocarbon samples from the

15

Marial site on the Rogue River within the current project
boundaries .

The teirmination of the Marial Phase and the beginning of
subsequent Coquille Phase is presently ill-defined, but has been
provisionally set at 4,500 BP. The Coquille Phase, which lasted
until approximately 2,200 years ago, is defined primarily on the
basis of a continued decrease in the size of leaf-shaped
projectile points and the appearance of a distinctive dart point
with a broad neck width made from locally available stone
material (jasper and chert) . This Coquille dart point does not
appear in collections in neighboring regions, and according to
Pettigrew (1990:67), may reflect the immigration of Athapaskan-
speakers into the region. Other traits of this era include a
dramatic decrease in the use of obsidian as a tool stone, but the
continued presence of a large number of end-scrapers.

The terminal era of this chronology is the Rogue Phase,
dating from 2,200 BP to contact with Euro-Americans. The Rogue
Phase is associated with the introduction of the bow and arrow
into the region and the parallel development of smaller, narrow-
necked arrow points. End-scrapers decrease in number, and the
use of obsidian remains at a low level (Pettigrew and Lebow
1987:11.61) . A utilitarian form of fired-clay, ceramic vessels
and figurines have been recovered from the excavations of village
sites on the upper Rogue and Klamath Rivers, dating from the
middle years of this phase. It is also during the latter 1,000
years of this period that permanent villages were established in
the Elk Creek drainage, although Pettigrew and Lebow (1987)
suggest that semi -subterranean housepits, and an associated
cultural pattern of semi -permanent residence, may have a much
longer history in the region, perhaps extending back as far as
the Marial Phase in the Applegate River valley.

The Forager/Collector Model: In this model of land-use and food-
gathering patterns, Winthrop (1993) looks at the two different
ways of life during the Archaic era; the early mobile, forager
pattern, and the later sedentary, collector pattern. Due to a
lack of archaeological data relating to the early Archaic era,
Winthrop' s model focuses on the Middle Archaic (7,000 BP to 2,000
BP) and the Late Archaic (2,000 BP to 150 BP) . The difference
between the two patterns is that during the earlier period people
moved among resources, using them as they became available and
living in temporary shelters; in the later period, collectors
emphasized the processing and storage of foods at permanent home
bases (i.e., villages). The intent of this research was to
demonstrate the shift (and when it occurred) from the long-lived
mobile, foraging pattern of settlement and subsistence to the
more sedentary collector pattern.

Winthrop analyzed the archaeological information from 83
sites in southwestern Oregon (43 from the Rogue River basin and

16

40 from the Umpqua River basin) , sorting the data into
chronological periods (Middle or Late Archaic) . The artifact
assemblages were also sorted (based on a number of criteria) into
one of three functional categories: village sites, seasonal
basecamp sites, or temporary, specialized tasks sites. The
operating premise of this functional classification was that the
housing and land-use patterns of the mobile forgers would be
reflected only in seasonal camp sites and specialized tasks
sites, while the more sedentary collector pattern would be
expressed in village sites, in addition to continued use of
upland seasonal camp sites and specialized tasks sites.

By comparing the functional site types with the
chronological data (projectile point styles, radiocarbon dates,
etc.) Winthrop was able to show that there was a correlation of
site types with time periods, thus establishing that the mobile,
forager pattern was the dominant settlement and subsistence
pattern during the bulk of the Middle Archaic era. Through
analysis of the data from the Rogue River basin, she postulates
that the shift from the forager pattern to a collector regime
probably began around 3,000 BP along the main stem of the Rogue
River, and spread throughout the region during the last 2,000
years (Winthrop 1993:208). Although people still used seasonal
upland base camps and specialized task sites (see the
ethnographic period discussion below) , the land-use patterns in
the Late Archaic were tied to permanent village sites. During
this period people invested time in substantial architecture
(i.e., pithouses) ; land management practices (e.g., controlled
burning to improve game forage and plant production) ; and more
intensively utilized, processed, and stored the plant and animal
resources within their home territory.

The reasons for this shift from a forager to a collector
pattern are not entirely evident. Although environmental data
relating specifically to southwestern Oregon during the Archaic
era is lacking, changes in the climate perhaps played a
significant role. As discussed in the previous chapter on the
environment, the climate of much of the Middle Archaic period in
region is believed to have been much drier and warmer than at
present. The drought conditions began to abate towards the end
of the Middle Archaic, when the collector pattern makes it
initial appearance. This change to a cooler and wetter climate
by 3,000 BP probably made the interior valleys of the region much
more productive in terms of plant resources, as well as in the
dependability of seasonal fish runs. These changes perhaps
encouraged the development of a more sedentary, collector
lifestyle along the principal rivers of the region. Natural
population growth, as well as cultural ideas brought by the
arriving Athapaskans, may also have stimulated and accelerated
the shift that is seen by Winthrop to have originally developed
with the resident population of the region.

17

FIGURE 2.
CHRONOLOGICAL PERIODS IN SOUTHWESTERN OREGON

Regional
Patterns
(Connolly 1986)

Regional

Chronology

(Pettigrew &

Lebow 1987)

Forager /Col lee tor

Model
(Winthrop 1993)

Years
BP

Gunther

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000 . . .

10,000 ;.

11,000

12,000

Siskiyou

Glade Tradtion

Rogue Phase

Coquille Phase

Marial Phase

Applegate Phase

Paleo- Indian

Late Archaic

Middle Archaic

Early Archaic

18

Issues in Regional Archaeology

The archaeological record of southwestern Oregon presents a
number of conflicting issues that await resolution. At this
point in our understanding of native cultures of the region, the
following research issues are perhaps the most pressing.

1. Environmental Change: A reliable regional model of
climatic and environmental change over the past 10,000 years
is badly needed. Hunting/gathering societies were
intimately connected with the natural environment. The
availability of food resources and the seasonal
accessibility of those resources were significant factors in
the survival strategies of the region's native inhabitants.
A clearer definition of past environments and a knowledge of
the time frame of climate change are necessary to understand
the economic and social life of native societies.

2. Stylistic Dating: Due to the paucity of radiocarbon
dates (especially in the Early and Middle Archaic periods)
and the lack of data from well stratified archaeological
sites, cross-comparison of projectile point forms has played
a major role in dating sites in the region. However, this
technique is not without controversy. As noted in the above
discussion concerning region-wide cultural patterns and a
regional chronology, there is a lively debate between those
who see a long record of stability, with only slight changes
in point styles (cf . Connolly) and those who detect a much
greater degree of variability in projectile point styles
that can be tied with specific time periods (cf. Pettigrew
and Lebow) .

3. Cultural Change: The timing and causes for the shift
from a mobile, forager way-of-life to the collector pattern
center on permanent villages. There is a general consensus
among researchers that both of these patterns are
represented in the archaeological record in southwestern
Oregon. However, the timing of shift, the reasons why, and
the exact nature of each of the two lifeways is still in
doubt. Pettigrew and Lebow (1987:10.22) believe this
pattern may have begun quite early during the Marial Phase,
ca 5,000 BP, as evidenced by housepits in the Applegate
River drainage and similar developments at about the same
time in surrounding regions. Connolly (1986), on the other
hand, doesn't see the change until much later, beginning
around 1,700 BP. And Winthrop (1993) strikes a position
mid-way, with the development of village life beginning
possibly as early as 3,000 BP along the Rogue River.

The causes associated with this fundamental shift in
the settlement and subsistence patterns are also under
scrutiny. Was the collector regime developed by the long-

19

time inhabitants of the region as postulated by Pettigrew
and Lebow? Or were outside influences and/or the influx of
newcomers to the region responsible for the change, as
advanced by Connolly? Perhaps a combination of these two
schools of thought is closer to the truth. How important
were environmental factors? Did increases in population
necessitate a new economic and social order? The answers to
these questions will undoubtedly raise new ones as
archaeological research in the region continues.

4. Athapaskan Migration: By the time of the arrival of the
first Euro-Americans in the region the Athapaskan- speakers
were well established on the southern Oregon and northern
California coast and inland as far as the Applegate River.
Linguistic studies have dated their arrival in the area to
between 1,500 and 1,000 BP. However, the route they
traversed from the original homeland in western Canada and
Alaska to southern Oregon, either along the coast or inland
and then over the Cascades, is not yet resolved.
Archaeologically, are they associated with the onset of the
collector settlement pattern and the appearance of the bow
and arrow? What was the nature of their influence on the
existing, resident population of the area? Did they
displace other native societies who had inhabited the land
before them?

The archaeological resources within the current project
area along the Rogue River may hold the key to many of these
questions concerning the Athapaskans . This stretch of the
Rogue River was the borderland between the Athapaskans and
the Penutian- speaking Takelma, who are believed to have
resided in southwestern Oregon for at least 5,000 years
(Aikens 1993) .

20

IV. THE ETHNOGRAPHIC RECORD
Introduction

Although what we know of the native way-of-life prior to the
arrival of Euro-Americans is, at best, fragmentary and based on
the memories of only a few elderly informants, those memories
have provided an outline of the seasonal subsistence patterns,
insights into the religious and ceremonial practices, snapshots
of daily life, and an understanding of many of the tools used in
everyday, routine tasks. With some caution, archaeologists can
project those same patterns and lifeways back in time to help
explain the archaeological record. Caution is needed because the
lifeways captured in the ethnographic accounts and historical
documents reflected only a moment in time. Change is the one
constant in the history of human development; what was true in
1840 was probably quite different 1,000, 5,000 or 10,000 years
earlier.

The Takelma

The ethnographic record for interior southwestern Oregon is
limited due to the rapid displacement of the Native American
cultures of the region after the "Rogue Indian Wars" of the
1850s. Early in this century Edward Sapir (1907) undertook the
first research on the Takelma Indians of southwestern Oregon; he
was followed in the 1930s and 1940s by John Peabody Harrington
(1981) , Philip Drucker (1940) , and Homer Barnett (1937) . Their
conversations with Francis Johnson (a Lowland Takelma) and Molly
Orton (an Upland Takelma) have provided us with glimpse of
traditional Takelma life. The writings and field notes of these
and other ethnographers have recently been synthesized by Gray
(1987) . This ethnographic synthesis is the basis for the
following description of the Takelma Indians who inhabited the
region in the vicinity of the current project area before the
arrival of Euro-Americans.

In the late prehistoric period the Takelma, a Penutian-
speaking people, resided in a territory that centered on the
upper and middle stretches of the Rogue River and extended east
up Little Butte Creek to the crest of the Cascades. To the
south, they occupied portions of the Bear Creek Valley as far as
the Talent /Ashland area; on the west, the Applegate River Valley
and Galice Creek marked the boundaries with their Athapascan-
speaking neighbors, the Dakubetede and the Taltuctuntede . The
Athapaskan Shasta Costa group resided along the banks of the
Rogue River, also to the west of the Takelma, below Grave Creek.
The Hokan- speaking Shasta shared the southern portion of the Bear
Creek Valley with the Takelma. Shasta territory extended south
and east into northern California along the Klamath, Shasta, and
Scott Rivers (Holt 1946) . The groups bordering Takelma territory

21

to the north were the Molala and the Cow Creek Band of Umpqua
Indians .

Frances Johnson
Takelma Informant

Courtesy of
American Philosophical Society

The Takelma, as defined by language dialect, were divided
into two, and possibly three distinct groups. The principal
villages of the Lowland Takelma were centered around the Rogue
River, extending from the present-day town of Gold Hill downriver
to perhaps Grave Creek. The Upland Takelma winter village home
territory was further upriver in the lower Bear Creek Valley near
Table Rock and perhaps as far east as Ashland, Oregon. The
drainage of Little Butte Creek was also considered Upland Takelma
territory. A third dialect group of Takelma may have inhabited
the upper reaches of the Rogue River drainage in the vicinity of
Trail and Elk Creek, although little is known of this sub-group
(see Figure 3) . All of the Takelma shared a common way-of-life
and a similar natural environment, though local differences in
the availability of certain resources may have resulted in
slightly different subsistence and settlement patterns.

The staple plant foods of the Takelma were acorns and camas .
A variety of root crops, manzanita berries, pine nuts, tarweed
seeds, wild plums, and sunflowers augmented their diet. Acorns
from the black oak were preferred over others, and were collected
in the fall of the year. The acorns were shelled and then

22

pounded into meal by means of a pestle used with a hopper-mortar
base (a bottomless basket placed on a flat rock) . The meal was
then leached with hot water over a clean bed of sand to remove
the bitter taste from the acorn prior to boiling the meal with
hot stones in a tightly woven basket. The camas bulb, of which
there were several varieties, was collected in flat, well-watered
areas adjacent to the Rogue River. The camas bulbs were
harvested with the aid of a digging stick made of mountain
mahogany. The bulbs were cooked in a rock- lined earth oven, then
mashed into a dough, formed into cakes, and stored for winter
use .

Anadromous fish (especially salmon) , deer and elk, as well
as a variety of small mammals (e.g., rabbit and squirrel) and
certain insects provided protein in the Takelma diet . People
fished during seasonal spawning runs that occurred in the summer,
winter, and spring for different species of salmon, although not
every fish-bearing stream had runs of fish every season.
Favorite fishing locations were at falls and rapids along the
Rogue and its tributaries. Fish were taken with dip-nets, hook
and line, and spears. Salmon were skinned and split through the
backbone; the head, tail and entrails were removed, and the meat
roasted on spits of split hazel branches. Cooked salmon was then
pulverized and stored for winter use. The Upland Takelma hunted
deer in several ways, including stalking with a deer head
disguise, running down deer in winter snows, snaring, and using
dogs to drive deer into fences set with nooses. The Lowland
Takelma seemed to have been more dependant upon riverine
resources and would often trade salmon for deer hides and meat
with their Upland Takelma kin.

The settlement pattern of the ethnographic-era Takelma was
closely related to their subsistence regime. The permanent
winter villages were located in the low elevation river valleys
of the region in close proximity to predictable and significant
food resources (i.e., near good fishing locations and/or root
crop fields and acorn groves) . The winter villages provided
storage for surplus food goods, and served as a firm geographic
locus from which the inhabitants would range out to secure other
resources on a fairly predictable seasonal basis. During the
warmer months of the year the Takelma would temporarily move to
seasonal base camps in the surrounding uplands to hunt, to gather
crops as they ripened during the summer season, and to procure
other resources not available near their winter villages (e.g.,
lithic raw material for the manufacture of chipped-stone tools) .

23

Dipnetting for Salmon (Artist's Conception)

The winter dwellings of the Takelma were described as
rectangular in shape and semi- subterranean to a depth of between
18 and 24 inches. The framing consisted of upright posts in the
corners, connected by cross beams to which exterior siding was
attached. The siding was either of split sugar pine planks
(reserved for the wealthier class) or with slabs of bark (for the
less fortunate and also ascribed to members of the Upland Takelma
group) . The roofs were of a single gable design covered with
planks and sometimes with earth. A fire hearth was located in
the center of the structure with a corresponding smoke hole in
the center of the roof. Kroeber (1953:905) gives the exterior
dimensions as being 12 feet wide and 15 to 20 feet long, with no
interior partitions. During the warmer months of the year, when

24

staying at upland base camps, the Takelma constructed temporary
shelters of brush or boughs around a central fire pit. The
Takelma also built sweathouses, generally one to a village. The
men's sweathouse was rectangular in shape, semi -subterranean, and
earth covered. The women's sweathouse was smaller and temporary,
consisting of a stick frame covered with woven mats.

Housepit Village (Artist' s Conception)

The plants and animals exploited by the Takelma not only
furnished food, but also the raw materials from which to make the
daily necessities of life. Buckskin was used to make shirts,
leggings, moccasins, and blankets, while elkhide was fashioned
into protective body armor for warriors in battle. A multitude
of basket styles were made from the roots of sugar pines; they
held everything from babies to boiling water. Baskets were

25

indispensable in the gathering of vegetable crops and were also
used for food storage. The ethnographic record is silent on the
subject of clay pottery, even though an expedient form of pottery
termed Siskiyou Utility Ware has been recovered in a number of
late prehistoric sites along the Rogue and Klamath Rivers. A few
other articles of everyday use are mentioned in the various
accounts of domestic life. Needles were made of hard wood or
bone; thread from sinew. Combs were made from a split stick into
that were inserted porcupine quills for teeth. Pipes of wood or
stone were used for smoking tobacco (the only plant cultivated by
the Takelma) , and string and rope were made from the iris plant.
Spoons were fashioned either from wood or elkhorn; small wooden
paddles were used to stir cooking foods.

The seasonal and annual fluctuations of the staple food
sources of the Takelma, combined with need to gather widely
scattered plant and animal foods in the upland areas, served to
isolate families and communities at various times. These factors
restricted the development of any strong central authority in the
region. Instead, the local village community was the principal
social and political unit. Ranking by wealth was a
characteristic of Takelma society. The possession of exotic
goods such as large obsidian blades and dentalia shells enhanced
status among the kin groups of the Upland Takelma. A position of
wealth in Takelma society carried with it certain
responsibilities as well as privileges. A wealthy man was
expected to support and benefit his poorer relations in times of
economic distress, in addition to providing for their marriages
and legal obligations. In return, a wealthy individual would
receive the labor of his beneficiaries in food gathering
activities and domestic chores, as well as their support in times
of political turmoil. Legal disputes were settled on the
principle of demanding compensation for injuries sustained; up to
and including murder. The services of a go-between were retained
by the aggrieved parties to facilitate the negotiation process.

The religious and ceremonial aspects of Takelma life are not
easily separated from the more mundane daily subsistence tasks.
In the world view of the Takelma, the forces of nature and the
fate of humans were determined by supernatural spirits. Many of
these supernaturals were associated with organic elements such as
plants and animals that were believed to be the transformed
manifestations of primeval earthly inhabitants. On the local
level, some supernatural beings were directly associated with
particular natural objects, including rocks, trees, and
mountains. It was to these specific objects that offerings of
food, valuables, and prayers were often made. Subsistence
occasions of importance for the Takelma included a celebration of
the new acorn crop, as well as a first- salmon ceremony. The
puberty rites for young women are well documented, and were an
important ceremony and celebration practiced within the
community. Upon the death of a member of the village, the body

26

was washed, painted, dressed, and flexed, prior to its removal
from the dwelling through an opening in the wall or roof. A male
relative would carry the body to the grave site, where it was
interred in a small, oval -shaped grave. The mourners placed
items of value (chipped- stone blades, dentalia, etc.) into the
grave .

A powerful figure in Takelma society was the shaman (Goyo) .
He or she (for both men and women were shamans) was deemed
responsible for causing illness and death, as well as preventing
the same. Additionally, shamans were employed to intercede in
the affairs of nature, for example by inducing rain in time of
drought or halting a severe winter storm. Shamans were generally
distrusted, so much so that some villages would not permit their
residence. Shamans obtained power from one or more guardian
spirits. In affecting a cure on an ill patient, the shaman would
communicate with his or her guardian spirits through song and
dance, aided by several helpers. The guardian spirits would
direct the shaman to the location of the "pain" that was then
physically removed by the shaman. Success of the operation would
depend upon the power of the shaman; the use of more than one
shaman on a specific case was not unknown.

In addition to shamans, another class of healer operated
within Takelma society, the S-omloholxa' s . They differed from
the Goyo in several ways. The S-omloholxa' s appealed to a
different set of guardian spirits; they could cure illness, but
not inflict it; they did not dance or require the services of
helpers in singing the medicine song; and they cured a patient by
rubbing the afflicted part of the body rather than "catching the
pain" . The limits of the S-omloholxa' s curative powers are
unclear, but it seems that in cases of serious illness, only the
spirit power of the shamans had a chance of success.

The Athapaskans

At the time of contact with Euro-Americans, Athapaskan-
speaking people inhabited the coastal areas of northern
California and southwestern Oregon, as well as inland territory
along the main stem of the Rogue River and its principal
tributaries, the Illinois River, the Applegate River, and Galice
Creek. The southern Oregon Athapaskan language is related to the
Na-Dene linguist family. This related group of languages is
found from Alaska and western Canada, where it is spoken by the
Tlingit and Haida groups, to the American Southwest and northern
Mexico, where it includes the Apache and the Navajo languages.

The Shasta Costa band of Tututni Indians occupied an area
along the lower Illinois River, as well as adjacent to the Rogue
River from Agness to Big Bend (Beckham 1978) . Within the project
area, they may have also inhabited the Rogue River Canyon as far
west as the western boundary of Takelma territory near Grave

27

Figure 3 Territorial Boundaries in Southwestern Oregon

28

Creek (Gray 19 87) . The Applegate Athapaskans, known as the Da-
ku-be-te-de, lived along nearly the entire course of the
Applegate River, from near its headwaters in the Siskiyou divide,
to its confluence with the Rogue River. The territorial extent
of the Gal ice Creek Athapaskans (the Tal-tuc-tun-tu-de) was
limited to the confluence of Galice Creek and the Rogue River, as
well as the surrounding area drained by Galice Creek (Gray 1987) .
The Shasta Costa Indians may have been quite numerous, as Dorsey
(1890) lists 33 villages within their territory. Only three
permanent villages were noted in the Applegate River drainage for
the Da-ku-be-te-de (Jacobs n.d.), none of which were on the Rogue
River. The principal village of the Galice Creek Athapaskans
(Talda' cdan) was located on both sides of the Rogue River at the
mouth of Galice Creek.

Canoe on the Rogue (Artist's Conception)

29

The material culture (i.e., tools, clothing, architecture,
etc.) and the subsistence economy of the Athapaskan- speakers of
southwestern Oregon, was very similar to that described above for
their neighbors, the Takelma. One ethnographer (Drucker 1940)
stated that the Galice Creek culture was "...so permeated with
Takelma elements as to be scarcely distinguishable from the
culture of these alien people." Whether in fact it was the
Athapaskan culture that was permeated with Takelma elements, or
the other way around, has yet to be determined. However, the
point is that for the most part, the native inhabitants of the
region shared a common way-of-life in a similar environment.
Groups were distinguished from one another by their language,
their historical traditions, and by different subsistence regimes
that correlated with the food resources that were available to
them within their home territories.

One of the more significant differences between the
Athapaskans and their Takelma neighbors revolved around religious
figures. Like the Takelma, the Athapaskans (at least those
dwelling on Galice Creek) had two classes of shamans. One,
similar to the Takelma Goyo, healed through the power of their
guardian spirit with a combination of herbal medicines, chants,
dances, and sleight-of-hand. The other form of religious figure,
quite different from the Takelma S-omloholxa' s , was the
formulist. The primary duty of the formulist was the formal
recitation of appropriate ritual words in all situations that
involved a formal religious procedure (e.g., first salmon rites,
life crises, etc.). Formulists gained their power through a
knowledge of formulas; they acquired their knowledge through
learning, as opposed to shamans, whose power was revealed through
the supernatural. The formulists differed from the Takelma
S-omloholxa' s in that they were more powerful than the Galice
shamans and they had the ability to inflict suffering on their
fellow humans. The Galice formulists were also more powerful
than shamans and were sometimes used as instruments of revenge
against personal enemies.

SO

V. NATIVE AMERICANS AND THE RIVER
Introduction

Archaeological investigations within the current project
boundaries, although limited in scope, have had a significant
influence on our understanding of the cultures that occupied the
Rogue River basin over the past ten thousand years. Four major
archaeological sites have been partially excavated in or near the
Rogue River scenic corridor. These include the sites at Marial
and Stratton Creek, the Marthaller site, and the Ritsch site. In
addition, a recent review of pertinent historical documents, and
a field inventory (Gray 1994 and 1995) of a sample of known or
suspected archaeological sites, augments the ethnographic record.
These sources provide time depth to the way-of-life described in
the last chapter.

In order to present the archaeological record of the native
cultures who lived along the Rogue River we will use the
generalized chronological framework employed by Winthrop (1993) .
This framework begins with the Paleo- Indian period, then divides
the longest time span, the Archaic era, into an Early, Middle,
and Late sequence. Within the span of the human occupation of
the Rogue River region, the four sites mentioned above serve as
examples for the various eras. The information from these sites
illustrates the material remains from different time periods, and
helps assess some of the issues that have been raised by the
cultural/chronological models reviewed in Chapter III.

It has only been in the last twenty years that
archaeologists have begun to investigate the remains of native
cultures along the Rogue River within the project area.
Beginning in the late 1970' s with the excavation of the Ritsch
site near the mouth of the Applegate River, excavations continued
through the decade of the 80' s with work at the nearby Marthaller
site and with a series of excavations downriver at Marial. More
recently, investigations were undertaken at the Stratton Creek
site. The location of these sites spans the length of the
project area, giving us a geographically representative sample of
habitation sites along the river (Figure 4) . These four sites
also represent at least eight thousand years of human occupation
for this stretch of the Rogue River. The sites at Marial and
Stratton Creek were occupied repeatedly during the entire course
of the Archaic era. While the Marthaller site was perhaps
utilized during the Early and Middle Archaic, it, along with the
Ritsch site, provides solid evidence of occupation during the
Late Archaic.

The following discussion of the archaeology of the project
area draws on the techniques, concepts, and issues discussed in
previous chapters .

31

to Portland

O R EGON

^

▲ ARCHAEOLOGICAL SITES

1

Ritsch

1

2

Marthaller

1

3

Stratton Creek

J

J

4

Marial

LEGEND

.. . . • Wild River
ooooo Scenic River
Recreational River

Figure 4 Location of the Excavated Archaeological Sites within
the Project Boundaries

32

Paleo- Indians (12,0Q0 BP to 10,500 BP)

This time period is the most poorly represented era, not
only along the Rogue River, but in southwestern Oregon as a
whole. Only a few isolated Clovis points have been located in
the region, and those have been dated solely by the cross-
comparison with similar, dated points from outside of the area.
At this point in our understanding of the Paleo- Indian culture,
it is probably safe to assume that the nomadic big-game hunting
way-of-life existed somewhere within the project area. However,
due to the changing nature of the river as it alternately erodes
and builds the landscape along its path, the chances of
documenting this earliest of eras are slim.

Recent excavations at the Stratton Creek site (Ross and
Blalack 1994) within the project boundaries have offered a
tantalizing clue to this Paleo- Indian epoch along the river. At
approximately two meters (six feet) below the surface, the broken
base of what may be an obsidian spear point of the Paleo- Indian
era was recovered. Although not a classic Clovis spear point,
the size and shape of the fragment indicates it may date from
this early period. Unfortunately, no radiocarbon dates are
associated with this possibly ancient tool. Whether this find
truly indicates the presence of the Paleo- Indian tradition along
the Rogue River is debatable; at the present time it is all we
have .

3 cm

"Paleo" Projectile Point Base from Stratton Creek
(Ross and Blalack 1992)

33

The Early Archaic (10,500 BP to 7,000 BP)

During the early centuries of the Early Archaic, evidence of
human activity along the Rogue River is nearly as scarce as it
was during the Paleo- Indian times. From 10,500 to 8,500 BP (what
has been termed the Applegate Phase of the Rogue Valley
chronology) , the archaeology in the project area lies almost
silent. Perhaps the point fragment from the lower levels of the
Stratton Creek site may relate to this era, or there may exist
more deeply buried artifacts at Marial . However, firm
indications of this early time period in the region along the
river remain illusive.

The archaeological site at Marial at the extreme western end
of the current project area, provides us with not only the
earliest, firmly dated site in southwestern Oregon, but also a
site that was occupied time and time again down through the
centuries into the Late Archaic. The information potential of
the archaeological remains at Marial are truly monumental . As of
today, only a superficial glimpse of life at Marial over the past
8,000 to 9,000 years has been uncovered.

Before describing the contribution of the Marial site to our
understanding of the Early Archaic period, it is necessary to
digress from the chronological format in order to explain the
cultural deposits that have been discovered at this important
site. A brief description of the history of excavation at the
site and importance of the information gained thus far will help
to place the Marial site in its proper context.

The archaeological deposits at Marial, located on lands
managed by the Bureau of Land Management near the mouth of Mule
Creek, had been the object of vandals during the 1970' s. In 1978
the BLM conducted limited testing of a portion of the site in
order to determine the nature of the cultural materials present
and the extent of the damage caused by illegal digging.
Subsequently the site was protected with heavy wire mesh buried
in the sod over the site (Deich 1983:2) . In 1982 the BLM, in
conjunction with volunteers, conducted more extensive
archaeological testing at the site. Seventy square meters of the
site were excavated in 19 test units spread across the site.
Excavations reached a depth of more than three meters, and
revealed the presence of two distinct stratigraphic levels. The
upper cultural level (from the surface to 90 centimeters below
the surface) contained artifacts (small arrow points and larger
side-notched and shouldered dart points) associated with the
Middle and Late Archaic periods; the lower cultural deposit was
radiocarbon dated to 6485 BP and yielded a variety of leaf-shaped
dart points. A large number of end-scrapers, as well as drills,
knives, ground- and battered- stone cobble tools and nearly 10,000
waste flakes from the manufacture of chipped- stone tools were
also recovered during the excavation.

34

Excavation at Marial 1983

Due to the potential significance of the Marial site (i.e.,
the early radiocarbon date and the existence of distinct cultural
layers), further excavation was planned for the summer of 1983.
This time, Oregon State University was the lead agency, assisted
by the BLM and volunteer organizations. The 1983 excavations
confirmed the presence of the two cultural levels as postulated
in 1982 and determined that cultural material was present to a
depth of at least 350 centimeters; a meter deeper than the
radiocarbon date of 6485 BP . This season of field work also
defined a Late Archaic occupation area, characterized by small,
barbed arrow points, south of the older deposits along the edge
of the terrace above the Rogue River.

Oregon State University returned again in the summer of 1984
for a more in-depth excavation of the older deposits in the
central portion of the terrace. The test units of the previous
year were reopened and expanded. A total of seven distinct
stratigraphic occupations, six of which were in the central
portion of the terrace, were defined. Most of these occupational
levels were easily distinguished within the excavation units; the
dark, organic soils containing large numbers of artifacts were
separated by layers of relatively artifact-free, light-colored
sand (Figure 5) . During the 1984 season four additional
radiocarbon samples were collected from different levels
throughout the excavation. The earliest radiocarbon date of
8,560 BP came from below Cultural Zone 6 at a depth of 430

35

centimeters (14 feet) below the surface; a few artifacts were
recovered below the radiocarbon sample. The excavation
terminated at 450 centimeters below the surface on top of a
gravel deposit.

4- 4^ , -^ -^

dark grayish brown

XVI

2810±50BP»
XV Cultural Zone 1

XIV

XIII

XII Cultural Zone 2
4060 ± 55 BP»

5850 ± 120 BPI

XI Cultural Zone 3

6485 ± 80 BPI

X

IX Cultural Zone 4

VIII

VII Cultural Zone 5

VI

V

IV

III Cultural Zone 6

8560 ± 190 BPI

Figure 5 Idealized Profile of the Marial Stratigraphy
Aikens 1993

36

The 1984 excavations greatly expanded the artifact
collection from the site, in addition to firmly dating most of
the occupation levels. The emphasis of the researchers at this
phase of investigation was to build a reliable projectile point
chronology for the site, tied to radiocarbon dates and
stratigraphic levels. Brief descriptions of other artifacts and
associated cultural material (e.g., animal bone fragments) were
presented in the final reports.

In 1985 OSU and volunteers returned briefly to Marial to
test the Late Archaic deposits along the terrace edge, south of
the major portion of the site. A radiocarbon date of 710 BP was
obtained from a fire hearth uncovered 1.5 meters below the
surface. The recovered artifacts (i.e., projectile point styles)
appear related to the Late Archaic era (Clark 1988) .

Returning again to the Early Archaic era along the Rogue
River, the data recovered from the lower levels of the Marial
excavations (Cultural Zones 4, 5, & 6) gives us the first glimpse
into the lives of the people who lived at the mouth of Mule
Creek. These three levels represent the time period from 8,560
BP to approximately 7,000 BP (a radiocarbon date of 6,486 BP was
obtained from the lower level of Cultural Zone 3) . Cultural
material recovered from these levels includes leaf- shaped and
broad- stemmed/shouldered dart points, end- scrapers, cobble tools
used in the manufacture of chipped- stone tools, waste flakes from
stone tool manufacture, and a small number of animal bone
fragments. No cobble tools related to plant processing (e.g.,
pestles or grinding slabs) were recovered from these levels, nor
were any features or indications of habitation structures noted.
It should be kept in mind, however, that the total volume
excavated in these levels was considerably less than in the upper
cultural zones. The great depth of the excavation (over 14 feet
deep) necessitated stair-stepping the walls of the excavation
units to prevent cave- ins. Thus, the lack of features and some
classes of tools in these lower levels may be due to the limited
sample size.

The cultural deposits unearthed and the artifacts recovered
from the Early Archaic portion of the Marial site conform with
proposed models for settlement and subsistence of this era. The
presumed mobile, forager pattern is well represented by the
artifact assemblage of Cultural Zones 4, 5, & 6. Hunting and
butchering of upland game, as evidenced by the dart points, end-
scrapers, and animal bone fragments, appears to have been the
primary subsistence activity during these occupations.
Functionally related to these activities is the manufacture and
repair of chipped-stone tools, as represented by the waste flakes
and battered cobble tools. The artifact-bearing layers at this
depth of the site are relatively thin, compared with later
occupations at the site, suggesting either limited use of the
location, or low populations. This limited occupation, combined

37

with the lack of habitation or storage features, also supports
the theory that small mobile groups of hunters restricted their
use of any one area to relatively brief periods of time before
moving on to exploit resources at other locations.

If we look at the occupational history of the site in these
early centuries, another pattern of land-use related to
environmental factors is suggested. Each of the artifact-rich
layers at the site are separated by relatively sterile layers of
sand deposited during flood episodes. During the 1,500 years of
sporadic occupation represented by the lowest three Cultural
Zones, there were periods of time (perhaps hundreds of years)
when little, if any, human activity occurred at the site. After
the effects of a major flood, or years of annual flooding, it is
reasonable to assume that the favorable conditions that attracted
people to the terrace alongside Mule Creek would have been
diminished. The sandy soils that covered the terrace immediately
after a flood (s) probably inhibited the growth of vegetation
attractive to game animals, and thus to the hunters who pursued
them. The loss of one favorable hunting/camping site would not
have greatly affected these mobile foraging groups, for their
way-of-life was predicated on movement among numerous sites and a
variety of resources. Gradually however, a more favorable regime
of plant and animal life would have been reestablished, prompting
the return of foraging groups. Then, for generations, the site
at Marial served once again as a camping spot on the continual
round of food and resource gathering. This cycle of periods of
occupation followed by periods of abandonment was repeated
throughout the history of Marial and is one of the factors that
makes the site of such importance to archaeologists.

Supporting evidence for the mobile, forager way-of-life
along the Rogue River comes from another site excavated by Oregon
State University in 1992 . The Stratton Creek site is located on
a terrace above the Rogue River near the mouth of Stratton Creek,
approximately 31 miles upriver from Marial. Like Marial, the
Stratton Creek site had been the scene of repeated episodes of
illegal looting over the years. In 1989 the BLM, in cooperation
with the Rogue Valley Archaeological Society, conducted test
excavations to determine the dimensions of the cultural deposits
and the extent of damage caused by the looting. Test excavations
(Ottis 1991) indicated the presence of cultural material to a
depth of at least 150 centimeters (five feet) , and the presence
of artifacts that dated the site to the Middle Archaic period.
Based on these results, additional excavation at the site was
recommended .

During the summer of 1992, staff and students from Oregon
State, along with volunteers from the local area, conducted a
large scale excavation of the site. Cultural material was found
to a depth of 250 centimeters (eight feet) . Three cultural zones
(or strata) , based on soil type and artifact densities, were

38

identified during the excavation. The top 40 centimeters of the
site had been heavily mixed by historic farming activities and
illegal looting. Artifacts within this zone were deemed to have
lost much of their contextual integrity due to these
disturbances, and were lumped into one group designated as
Stratum I. Stratum II extended from 40 centimeters below the
surface to approximately one meter (three feet) below surface.
Both Stratum I and II were composed of relatively dark sandy
soil. Stratum III began at about one meter below the surface and
continued down to bedrock, nearly two and one-half meters below
the ground surface. The soils in Stratum III were lighter in
color and more compact than the preceding strata, and artifact
densities were much lower. No radiocarbon dates were obtained
during the course of the excavations, hence dating of the site
relied on the cross-comparison of projectile point styles with
the artifacts from the Marial site, as well as from a small
sample of obsidian hydration dates.

Stratton Creek Excavation, 1992

39

The Early Archaic period may be represented by the artifacts
recovered from lowest levels at the site within Stratum III.
According to the project researchers "comparative cultural
material suggests a date around 10,000 [BP] for the earliest
components (Ross and Blalack 1992)". Artifacts associated with
Stratum III included broad-neck, corner-notched and stemmed dart
points; leaf- shaped (lanceolate) projectile points; one point
possibly identified with the Paleo-Indian era (as discussed
above); heavy choppers made from river cobbles, hammerstones used
to make chipped-stone tools, and pieces of incised shale. The
incised shale pieces recovered from Stratum III are, at the
least, indicative of artistic expression; whether there was a
functional or religious significance to these pieces is unknown.

The artifact assemblage from Stratum III is similar in
function to that discussed for the lower levels of Marial .
Hunting and butchering of game (projectile points, scrapers, and
choppers) , and the manufacture of chipped-stone tools
(hammerstones and waste flakes) point to the use of the site as a
seasonal hunting camp. The lack of features or evidence for
substantial architecture, as well as the deficiency of tools
directly associated with the processing of plant foods (e.g.,
pestles or grinding slabs) also tend to suggest a short-term,
sporadic occupation of the site by mobile foragers during this
time period.

The obsidian studies on samples from the Stratton Creek site
have provided some interesting information. A total of 29 pieces
of obsidian (tools and waste flakes) from the lower two strata of
the site were analyzed as to their point of origin. Of those 29
samples, 20 artifacts originated from Spodue Mountain in the
northern Klamath Basin, 2 items were from Silver Lake/Sycan Marsh
(also in the northern Klamath Basin) , and 7 specimens were from
Grasshopper Flat in northeastern California. These obsidian
source locations have been noted at a number of sites throughout
southwestern Oregon, and indicate a fairly extensive trade/travel
network that existed for thousands of years.

Of the 29 samples submitted for sourcing analysis, a subset
of 20 from the Spodue Mountain source were also subjected to
obsidian hydration analysis. Fourteen specimens of this subset
came from Stratum III; six from Stratum II. The hydration
measurements for the Stratum III samples ranged from 3.6 microns
to 9 . 7 microns, with an average reading of 4.77 microns. The six
obsidian samples subjected to hydration analysis from Stratum II
returned readings ranging from 3.9 to 5.3 microns, with a mean of
4.56 microns; statistically similar results. These readings
appear to contradict the site stratigraphy (i.e., normally one
would expect significantly higher micron readings from Stratum
III than Stratum II) . Comparing these results with the absolute
dating sequence generated by obsidian studies in the Elk Creek

m

projects further upriver^, dates the majority of the obsidian
samples from Stratton Creek in the 5,000 BP to 6,000 BP range;
the early part of the Middle Archaic.

A closer examination of the hydration results and the
stratigraphy from Stratton Creek suggests some possible
resolutions to what appears to be contradictions among the
proposed projectile point chronology, the site stratigraphy, and
the obsidian dating results. Keeping in mind that a sample of 20
hydration readings is a very small sample from which to draw firm
conclusions regarding the dating of any site, the following
hypothesis are offered to explain the obsidian hydration data.

1 . Four readings were greater than 5 microns and probably
date from the Early Archaic. From Stratum III they were:
9.7 microns from level 16; 8.0 microns from level 15; and
5.4 microns from level 19. In Stratum II, a reading of 5.3
microns was returned from a tool fragment recovered in level
9 at the interface of Stratum II and III. This early
reading from Stratum II may indicated that this tool was
made from obsidian that had been brought to the site by
earlier inhabitants, discarded, and then reused at a later
date. The early hydration reading on this specimen probably
reflects the first use of the tool stone during the Early
Archaic, not in its later position in Stratum II.

2. In Stratum II, two other readings from level 8 near the
interface between Strata II and III produced rather high
readings of 4.7 and 4.9 microns. These samples appear to be
associated with a major rock feature (Feature 8) composed of
a pavement of river cobbles arranged in an oval pattern
sloping down towards the center, giving an impression of a
shallow depression (Ross and Blalack 1994) . The
construction of this rock feature may well have disturbed
the underlying, older cultural deposits, thus displacing the
obsidian from its earlier position. Alternatively, the
obsidian could have been displaced by the excavation of an
earlier (1991) test unit that was noted in the northwest
corner of 1992 unit (Ross and Blalack 1994) .

3. The remainder of the hydration readings (N=14) range
from 3.6 to 4.5 microns, with a mean of 4.01. This
translates to an average date of approximately 4,500 to
5,000 BP, solidly in the Middle Archaic period, or the

^ Using the absolute dating sequence from Elk Creek, which is
located at a higher elevation with colder average annual
temperatures, is less than ideal. However, the lack of radiocarbon
dates associated with the obsidian artifacts at the Stratton Creek
site leaves the Elk Creek data the geographically closest location
for obsidian dating comparison.

41

latter part of the Marial Phase of the regional chronology
as proposed by Pettigrew and Lebow (1987) . The fact that
the few (N=3) Stratum II readings are statistically the same
as the deeper Stratum III samples can perhaps be attributed
to mixing of the soils due to natural causes (rodent
activity and or root growth), or to sampling error.

In summary, the obsidian data from the Stratton Creek site
suggests that obsidian was imported from the northern Klamath
Basin and northeastern California during the Early and Middle
Archaic time periods. The majority of the obsidian tested from
the site dates to the Middle Archaic (ca. 4,500 to 5,000) . [No
samples from Stratum I were analyzed.] The few samples that have
readings greater than 5 microns tend to confirm the proposition
that the Stratton Creek site was occupied, at least sporadically,
during the Early Archaic. The obsidian data also indicates that
the strata that have been defined at the site on the basis of
soils and artifact densities may not directly relate to the site
chronology. In other words, the upper levels of Stratum III may
represent the early part of the Middle Arachic; while only the
lowest levels date to the Early Archaic. The lack of a distinct
stratigraphy within Stratum III was noted by the researchers.
There were subtle soil color changes recorded throughout Stratum
III; however, they were not correlated with distinct cultural
occupations. A larger sample of obsidian hydration dates from
each stratum, as well as a more detailed examination of
relationship between the site stratigraphy and recovered
artifacts, would greatly clarify the chronological issues at
Stratton Creek.

The Middle Archaic (7,000 BP to 2,000 BP)

A convenient environmental marker for the beginning of the
Middle Archaic in southwestern Oregon is the volcanic eruption of
Mt . Mazama, which is dated to approximately 6,800 BP . The
violence of the eruption of Mt . Mazama had a profound effect on
the landscape of the High Cascades, and deposited thick layers of
ash over regions to the north and east. On the west side of the
Cascades, the upper reaches of the Rogue and Umpqua Rivers were
choked with pumice and ash. As to the effects of this event on
the middle and lower stretches of the Rogue River within the
project area, little is known. Use of the Marial site may have
been abandoned for a short time after the eruption, although by
approximately 6,500 BP we once again find evidence of site
reoccupation.

The sites at Marial and Stratton Creek once again provide
evidence for human occupation along the river during this long
period of transition from the mobile, forager way-of-life to the
more sedentary existence of the collector pattern. The
transition from the Early Archaic to the Middle Archaic at 7,000
BP is a somewhat arbitrary division; no great technological,

42

economic, or social changes occur at that date. Rather, the rate
of change appears to accelerate during the Middle Archaic, as
compared with earlier times.

A radiocarbon date of 6,485 BP was obtained from a charcoal
sample from the lower portion of Cultural Zone 3 at Marial during
the 1982 BLM excavations, and its position was confirmed during
the 1983 OSU field season. Another radiocarbon date of 5,850 was
obtained from a sample located in the middle portion of the same
zone. Cultural Zone 3 is the thickest layer of cultural deposits
(approximately 50 centimeters) with the highest artifact density
recorded to date at the site. Both the number and variety of
artifacts recovered in Zone 3 mark a change in site usage during
this early period of the Middle Archaic. Although the
predominate projectile form consisted of a variety of leaf -shaped
dart points, the recovery of a number of other chipped- stone and
cobble tools suggests a more intensive occupation. In addition
to projectile points and end- scrapers, the number of tools made
from river cobbles reached its zenith in Zone 3 . These cobble
tools related not only to the manufacture of chipped-stone tools
(e.g., hammerstones and anvils), but also for the first time to
the processing of plant foods (e.g., grinding slabs and pestles) .
Other artifacts associated with this time period included the
distinctive McKee unifaces, incised shale pieces, gravers
(chipped-stone tools used to incise bone, wood, or soft stone) ,
an atlatl weight, a baked clay ball, a grooved cobble that may
have served as a fishing line sinker, as well as numerous small
pieces of animal bone.

Incised Shale From Stratton Creek (Ross and Blalack 1992)

43

Also uncovered within Zone 3 was the first evidence of
possible substantial architecture at the site. At 180
centimeters below the surface of the excavation a feature was
uncovered consisting of a cluster of river cobbles placed
together forming a solid round platform, a little less than a
meter (three feet) in diameter, and slightly concave in the
center. The researchers speculated, based on the presence of a
number of heat spalls split off from the river cobbles, that this
feature may represent the floor of a sweathouse . An ethnographic
account of Takelma sweathouses (Sapir 1907) documents that heat
for sweathouse was obtained by pouring water over heated stones.
This would account for the cobble spalls associated with the
feature. The radiocarbon date of 5,850 BP was obtained from a
sample near this feature (Schreindorfer 1985) .

Zone 2 , immediately above Zone 3 , was radiocarbon dated to
4,050 BP. This cultural layer was differentiated from the
previous zone by a new style of dart point, a stemmed/shoulder
form. All of these dart points were manufactured from local
materials (e.g., jasper or chert) and are similar in description
to the Coquille Series Corner-Notched points defined by Nilsson
and Kelly (1991) for the Elk Creek area. In addition to the
continued presence of cobble tools and animal bone fragments, one
large, sandstone pipe bowl was recovered at the interface of Zone
2 and 3 . The pipe bowl was decorated with etchings on the side
in a cross-hatched pattern (Schreindorfer 1985) . The use of
tobacco has been documented ethnographically for groups in the
region; the presence of the pipe bowl at this depth indicates
that this may have been a long-lived tradition.

Zone 1 may represent the transition from the Middle Archaic
to the Late Archaic eras dating from approximately 3,000 to 2,000
BP . This cultural layer was located below the historic plow zone
of the site, and extended to a depth of nearly a meter. It was
characterized by dark silty soils with the highest density of
fire-cracked rock for any of the zones at the site. Fire-cracked
rock results from the use of river cobbles in the preparation of
various cooked foods. One ethnographically documented method for
cooking foods involved the heating of stones in a fire hearth,
then transferring the heated stones to a basket. The heat from
the stones would then boil or cook whatever liquid or food was in
the basket. After using the stones in this manner, they would
eventually break into pieces too small to be reused. These
discarded cooking stones are referred to as fire-cracked rock in
archaeological sites. The densities of fire-cracked rock are
assumed to represent the intensity of site occupation over time.

Projectile point forms were the most varied in Zone I. They
consisted of a variety of stemmed/shouldered and leaf-shaped dart
points, as well as smaller corner-notched, side-notched and base-
notched arrow points. This admixture of dart and arrow points
suggests that Cultural Zone 1 was occupied for a significant

44

period of time before, during, and after the adoption of the bow
and arrow, generally assumed to have come into use in the area
around 2,000 year ago. A radiocarbon date of 2,810 BP was
obtained from a sample recovered in the lower half of the zone.
As with the previous occupation levels of the Middle Archaic,
animal bone fragments, as well as cobble tools relating to plant
processing and chipped- stone tool manufacture, were recovered. A
stone platform feature, similar to the one described for Zone 3,
was located near the bottom of Zone 1. This feature was
uncovered during the 1982 field season and was described as:
"... a roughly circular stone platform about a meter in diameter"
(Deich 1983) .

Projectile Points from Marial (Oldest Points at Bottom)

45

At Stratton Creek the lower strata of the site appears to
date primarily to the Middle Archaic era. Although no
radiocarbon dates were obtained from the site and the site
stratigraphy is less than clear, the artifact assemblage and the
obsidian hydration dates from Strata II and III at Stratton Creek
closely relates these deposits to the upper three cultural zones
at Marial . The predominant projectile point styles in Stratum II
at Stratton Creek are leaf -shaped dart point forms. The
researchers (Ross and Blalack 1994) identified 22 variants of
these leaf-shaped or lanceolate style projectile points. Several
of these forms persisted over a long period of time, while others
are only represented by a single example. Broad-necked, and
side- and corner-notched dart points, as well as
stemmed/shouldered dart points, similar to those recovered from
Marial, are well represented at Stratton Creek.

In addition to the various forms of projectile points, a
large number of artifacts related to the processing of game
animals {end-scrapers and cobble choppers) were recovered from
the site, along with gravers, chipped-stone drills and knives,
and cobble tools (i.e., hammerstones) used to manufacture other
stone tools. Several rock features were reported at the
interface of Stratum II and III. Although most of the rock
features had no discernable pattern and were not ascribed a
function by the excavators, one feature (#8) was compared to the
rock features uncovered at Marial . This feature was described as
"... flat, round river cobbles arranged in an oval pattern
sloping down towards the center, giving an impression of a
shallow depression (Ross and Blalack 1994) ." No speculation was
made as to its possible function.

Very few animal bone fragments were recovered during the
1992 excavations by OSU, however, during earlier testing in 1989
animal bone fragments were recovered from test units to a depth
of perhaps one meter (Ottis 1991) . The 1989 testing also noted
the dramatic decline in obsidian use that occurred over time at
the site. The ratio of obsidian waste flakes to flakes of local
materials (e.g., jasper or chert) was 1 to 6.5 in Stratum III,
but declined to 1 to 16 in Strata I and II. This change in the
use of stone material between the earlier and later periods has
been noted at other sites in the region (cf. Pettigrew and Lebow
1987) . The decline in obsidian use in the later periods at the
site is an intriguing question that remains to be answered.

Excavation reports from both 1989 and 1992 field seasons
document the large amount of fire- cracked rock that was
encountered in Strata I and II. The quantities of fire-cracked
rock fell dramatically in Stratum III, the oldest portion of the
site. Large quantities of fire-cracked rock are generally
associated with more intensive site occupation. Therefore, this
class of artifact supports the hypothesis that the site was used
only sporadically during the Early Archaic. Ground- stone

46

artifacts relating to plant processing (e.g., mortar stones and
pestles) were recovered only from the Late Archaic deposits in
the upper levels of Stratum II and within the disturbed deposits
of Stratum I, indicating that the later inhabitants at the site
had shifted to using a broader spectrum of available food
resources .

The archaeological remains related to the Middle Archaic
time period from the Marial and Stratton Creek sites confirm many
of the chronological and economic concepts proposed in the
regional models as set forth in Chapter III. Throughout the
5,000 years of the Middle Archaic, as represented by these two
sites, there appears to be a great deal of stability in the
cultural record. Both sites appear to have continued to function
as seasonal base camps for hunting and processing game, and
making and repairing stone tools. Evidence for the exploitation
of the Rogue River fisheries is almost non-existent. Site use
for processing plant foods only appears towards the end of the
era. The existence of the mobile, forager pattern is further
supported by the lack of evidence of substantial residential
structures (i.e., pithouses) , although the rock features found at
both sites hint that perhaps people were spending longer periods
of time at these locations. The archaeological evidence from
these two sites strongly supports the notion that hunters and
gatherers, for most of this period, continued to pursue a highly
mobile way-of-life, moving among the resources of the region.

In addition to cultural stability, cultural change is also
evident during the Middle Archaic. It may have been incremental
at times, but by the end of this era the stage had been set for a
rapid expansion into the collector, village way-of-life. New
technologies for processing and storing food were evidently being
refined during this era, and populations were increasing. A
larger population, and hence group cooperation and a more complex
social structure, together with developing technologies, laid the
foundations for a new economic order.

The increasing rate of change is well expressed in the
archaeological record at Marial and Stratton Creek in projectile
points forms. Even though the points were intended for the same
weapon systems (the spear, and dart and atlatl) point styles
changed over time. These changing projectile point styles may
reflect technological improvements in the weapon systems,
different point types used for different types of game or hunting
conditions, or merely stylistic fashions that changed with the
times. In any case, these changing artifact forms are an
indication of innovation and cultural adaptability.

Other changes noted in the archaeological record at Marial
and Stratton Creek towards the end the Middle Archaic are the use
of stone tools to process plant resources, an increase in the
variety of chipped- and cobble-stone tools, and higher densities

47

of fire-cracked rock. The changes in these artifact classes
argue for more frequent and perhaps longer periods of use of the
sites towards the end of the Middle Archaic, as well as a more
extensive use of available resources. People were beginning to
spend more of their time and energy at these sites, a precursor
to the collector pattern.

The trend is clear. The life of the mobile foragers was
beginning to change. It appears from the archaeological record
that the evolution towards the collector/village pattern
accelerated during the last two thousand years of the Middle
Archaic. Whether there were villages with substantial
architecture along the river during the latter part of Middle
Archaic as postulated by Pettigrew and Lebow (1987) and Winthrop
(1993) has yet to be proven, although research to date has been
extremely limited. What is clear, is that the stage was set by
the events of the late Middle Archaic for the change to a new
mode of existence based on different technological and economic
concepts; a new way-of-life was about to begin along the banks of
the Rogue .

The Late Archaic (2,000 BP to 250 BP)

Archaeologically, the beginning of the Late Archaic era in
southwestern Oregon is marked by the advent of the bow and arrow.
The exact timing of this technological innovation in weaponry is
a subject of debate, but a date of approximately 2,000 years ago
is close. The advent of this new weapon is indicated by smaller
and lighter projectile point forms, as compared with the heavier
projectile points used on thrusting spears or with the atlatl and
dart. The transition from the mobile forager subsistence pattern
to the more sedentary collector pattern intensified during the
Late Archaic. Other cultural hallmarks of this era include
expanded trade networks with both neighboring and distant groups,
and societies in which social distinctions were increasingly
marked by wealth. The migration of Athapaskan- speaking people
from northern latitudes to the coast of southern Oregon and
northern California, and inland to portions of the current
project area, took place during the Late Archaic. These events
transpired during a time when the climate and vegetation patterns
of southwestern Oregon were similar to those of today.

There are several excavated archaeological sites within the
project boundaries that were occupied during the Late Archaic.
However, one of the better documented excavations of a Late
Archaic occupation occurred along the Rogue River just east of
the upriver boundary of BLM jurisdiction, approximately a mile
east of the confluence of the Applegate and Rogue Rivers. The
Ritsch site was located during the course of a cultural resource
survey in response to a proposed sewer treatment project for the
city of Grants Pass. A preliminary test excavation at the site
revealed the presence of a buried housepit floor and associated

48

artifacts approximately one meter below the modern ground
surface. Subsequently, the site was partially excavated in late
1976 by students from Oregon State University, led by Dr. David
Brauner .

Two, and perhaps three components, or levels of occupation,
were identified at the site. Component I extended from the
surface down to a depth of approximately 50 centimeters. The
upper portion of this component was heavily disturbed by historic
farming, yet contained a substantial quantity of chipped- stone
and cobble tools, as well as fire-cracked rock. The lower
portion of Component I escaped the ravages of the plow. Taken
together these two layers of Component I were characterized by a
variety of chipped-stone tools, including scrapers, gravers,
projectile points, and waste flakes from tool manufacture. Tools
made from river cobbles were also recovered, such as
hammerstones, choppers, edge-worn cobbles and net-sinkers. The
presence of notched net-sinkers (evidence of fishing activity at
the site), was unique to this component of the site. Another
artifact unique to Component I was a form of concave-base,
triangular- shaped arrow point. Animal and fish bone fragments
were also recovered from this component. A radiocarbon date of
460 BP was obtained from a charcoal sample near the bottom of
Component I in an area that may have been a surface dwelling
(Wilson 1979) .

A layer of soil containing a relatively low density of
chipped- and cobble-stone tools extended for 50 centimeters below
Component I. Small base- and corner-notched arrow points were
common in this zone. The researchers believed that this portion
of the site represented "...light periodic use of the site
between the occupation of Component I and Component II and
accounts for approximately one thousand years of the sites 's
history (Wilson 1979)." Below this zone, at approximately one
meter below the surface, the floors of two semi -subterranean
housepits were encountered. This level of the excavation was
designated as Component II. One complete house floor was
excavated and nearly half of the other before the field season
was completed. In addition, the "living surface" area
surrounding the excavated house features was exposed by the
archaeologists. Radiocarbon dates taken from the fire hearths of
the houses dated them to 1,400 and 1,470 BP; these are
statistically equivalent dates. These are the oldest, firmly
dated housepits yet excavated in southwestern Oregon.

The floor of house number 1 was described by Wilson (1979)
as being circular in shape with a diameter of 3.5 meters (11.5
feet) . It had been excavated into the then existing ground
surface to the depth of 50 centimeters, with a central fire
hearth dug 10 centimeters into the central portion of the floor.
No evidence remained of the construction of the above ground
portion of the house. A few chipped-stone tools (projectile

49

points, a drill, a scraper, and some utilized waste flakes) were
located within the house fill. In addition, two hopper-mortar
bases, a number of cobble- choppers, an incised bone fragment, a
bone awl, two salmon vertebrae, as well as numerous pieces of
fire-cracked rock were associated with the floor of house number
1.

House number 2 was only partially excavated. It too was
circular, although somewhat larger; four meters in estimated
diameter. The depth of the house floor beneath the old ground
surface, and the placement of the fire hearth were similar to
house number 1. The artifact assemblage from house number 2 was
comparable to that of house number 1, although fewer formed stone
tools were recovered. No hopper-mortar bases or bone implements
were present. It was speculated that any useable artifacts had
been removed by the occupants prior to abandonment.

The area outside of the house floors yielded a greater
number of artifacts than the interior of the housepits. This
"living surface" extended from 100 to 130 centimeters below
modern ground surface, and encompassed an area between the two
house floors. A similar artifact assemblage of projectile points
and other chipped-stone tools, as well as tools crafted from
river cobbles were recovered from this area of the excavation. A
majority of the projectile points were small, triangular-blade,
corner- to base-notched arrow points. Quantities of fire-cracked
rock, as well as a few pieces of animal bone (deer) and shell
(river mussel) were noted by the excavators. A radiocarbon date
of 1,150 BP was obtained from this area that was described as a
"general or multi-use area (Wilson 1979) ."

Below the house floors and the "living surface" between
them, excavations uncovered cultural material to a depth of 1.9
meters below ground surface. Artifact densities were light and
the tool types similar to those noted above. At the bottom of
this cultural zone a circular fire hearth, with no associated
artifacts, was exposed. There was not sufficient charcoal
remaining within the hearth to obtain a radiocarbon date, nor
were any diagnostic projectile points recovered from this
stratum.

How early the first occupants originally used the Ritsch site
area is speculative; the recovered data from below Component II
is inconclusive. However, by 1,400 BP the site was being
actively occupied by people who invested a significant amount of
energy in constructing semi -subterranean house structures, and
who were exploiting a variety of locally available resources,
including plants, upland game, fresh-water mussels, and fish. By
approximately 500 years ago, a fishing industry using woven nets
(evidenced by net-sinkers) had been established at the site and
it overshadowed the hunting orientation of earlier times. The
presence of concave-base projectile forms in Component I led

50

Wilson (1979) to speculate that there was a strong affinity with,
or influence from, coastal groups with similar artifact forms.

Another Late Archaic site, the Marthaller site, lies along
the banks of the Rogue River less than two miles downriver from
the Ritsch site and west of the mouth of the Applegate River.
The Marthaller site is located within the scenic corridor managed
by the BLM. Illegal looting of the site prompted the BLM, in
conjunction with the volunteer group, the Oregon Archaeological
Society, to conduct a large scale testing of the site in 1981 to
evaluate the significance of the site and the extent of damage
caused by the ongoing vandalism.

A total of 42, five foot square units were excavated during
the 1981 field season. The depth of the cultural material
extended to approximately 140 centimeters below the modern ground
surface, although the bulk of the artifacts were recovered from
the top 90 centimeters. The chipped-stone tool assemblage from
the site was fairly extensive; tools were manufactured from
locally obtained lithic materials (chert and jasper) , as well as
from imported obsidian. Scrapers, drills, gravers, a large
number of waste flakes from stone tool manufacture, in addition
to various partially modified flakes and tool fragments were
recovered from the excavation. Over 150 identifiable projectile
points were analyzed as part of the project. Approximately 50
percent of the projectile points were deemed to be arrow points
dating to the Late Archaic (small basal- and side-notched points,
as well as triangular concave base forms) . It was primarily on
the basis of these arrow points, as well as a few other artifact
forms similar to coastal sites, that the researchers dated the
site to a period between 500 BP and contact with Euro-Americans
ca. 1827 (Steele 1984) .

A review of the description of the other 50 percent of
recovered projectile points suggests that perhaps the Marthaller
site may have been occupied for some time prior to 500 BP . A
large number of leaf -shaped projectile points and broad-necked
shouldered and stemmed points, generally assumed to be associated
with the use of the dart and atlatl, were noted in the final
report (Steele 1984) . Although the descriptions of these points
are somewhat generalized by today's standards, and their location
within the cultural deposits is not documented, many of the
specimens appear to fit the description of projectile points
assigned to the Coquille Phase (4,500 to 2,500 BP) , the Marial
(8,500 to 4,500 BP) , and possibly even the Applegate Phase
(10,500 to 8,500 BP) of the local regional chronology (cf.
Pettigrew and Lebow 1987; Nilsson and Kelly 1991);
Alternatively, these dart points may represent the continued use
of atlatl into the Late Archaic, or the recycling of old
projectile points. Unfortunately, no radiocarbon samples from
the site were submitted for analysis, nor was obsidian hydration

51

analysis available at that time. The lack of any clear cultural
stratigraphy also hinders the resolution of the site chronology.

Dart-Points from the Marthaller Site

Early Projectile Points from the Marthaller Site

That the Marthaller site was the location of a Late Archaic
settlement related to the collector/village pattern is not in
doubt. The presence of numerous grinding slabs, hopper-mortar
bases, and pestles represents a significant investment in the
manufacture of heavy plant processing tools. Other artifacts
attesting to a sedentary way-of-life include notched net-sinkers,
a stone pipe, soapstone bowl fragments, hammer and anvil stones,
polished and flaked cobble tools, and a number of abraders used
to straighten and smooth arrow or dart shafts. The shells of
fresh-water mussels and mammal bone (e.g., deer) fragments were
also present in the cultural deposits.

52

V. f

Net -sinkers from the Marthaller Site

Three fire hearths were discovered at the site during the
course of excavation; two were at a depth of approximately 60
centimeters, and another was at 90 centimeters below the surface.
No housepit features were recorded at the site. The lack of
housepits raises the possibility that the Marthaller site,
although apparently quite large and repeatedly occupied, may not
have been a classic semi -permanent winter village site like those
described in the ethnographic literature. Rather, the location
may have served as a more temporary, but heavily used, fishing
and plant gathering base camp. An account of the area recorded
by the ethnographer Melville Jacobs in the 1930' s from Native
American informants stated "...right at the mouth of the
Applegate there were a lot of people. That's where they were
dip-netting salmon. .. that time they lived in a summer [emphasis
added] camp" {Jacobs [nd] ) .

Stone Bowl Fragment from Marthaller Site

53

The other two excavated sites within the study area, Marial
and Stratton Creek, also provided evidence of use during the Late
Archaic. During the 1985 field season at Marial, the southern
portion of the occupied terrace was extensively sampled. The
artifacts recovered during this phase of the archaeological work
at the site, combined with limited testing of this portion of the
site during previous investigations, documents the human
occupation of the site during the past 1,000 years. The
projectile point forms recovered from the plow zone of the
central terrace, as well as from the southern portion of the
terrace are predominately small basal -notched arrow points
(Gunther-barbed) of the Late Archaic. Other chipped-stone tools
recovered included drills, scrapers, gravers, and fragments of
partially fashioned multi-purpose tools. A few cobble and
ground- stone tools, including shaft abraders, as well as a net-
sinker were also reported from the site (Clark 1988) . Numerous
small fragments of animal bone, in addition to a few fresh water
mussel shells were recovered during the 1985 excavations.

A radiocarbon date of 710 BP was obtained from a fire hearth
uncovered at a depth of 1.5 meters below the surface in the
southern part of the terrace (Ross 1987) , indicating that this
portion of the terrace has built up much more rapidly than the
center of the terrace where the older, stratified cultural
deposits of the Early and Middle Archaic periods are located.
This discontinuity between the location of the older deposits in
the center of the terrace and younger Late Archaic material on
the southern edge of the terrace has yet to be explained. In the
realm of speculation, it is possible that the Late Archaic
occupation at Marial represents a different cultural group (i.e.,
Athapaskan- speakers) , whose knowledge of the past use of the area
did not extend back for generations. Thus, their settlement may
have been predicated on functional needs (e.g., a heavier
reliance on fishing) rather than on historical or traditional
patterns .

To date, none of the investigations at Marial have
documented the presence of a permanent village site with
housepits. However, there is historical information that sheds
some light on this subject. The response to a question during an
interview (Atwood 1976) with Ivin Billings, a life-long resident
of Marial, may explain the lack of these features:

Question by Kay Atwood: "Did you dig up Indian artifacts in
this area [lower field] once in a while?"

Ivin Billings: "Yes... way down on the lower end... they
camped down there... the whole part of the lower end. The
pits were still there. We had to plow them up and rake dirt
in to smooth them up."

54

It appears that historic plowing (noted in the top levels of the
excavations) may well have destroyed any evidence of Late Archaic
housepits floors on the terrace.

The same fate may have befallen any evidence for housepits
at the Stratton Creek site. There, the top 40 centimeters of the
cultural deposit was heavily disturbed by farming. However, the
1992 excavations did recover a number of basal -notched arrow
points of Late Archaic vintage in the disturbed top stratum
(Stratum I), and in the uppermost levels of Stratum II. Stratum
I also contains dense quantities of fire-cracked rock, indicative
of intensive, repeated occupation. Whether the Stratton Creek
site served as a temporary base camp or as a semi -permanent
village during the Late Archaic is, at this point, difficult to
determine. Perhaps future investigations at the site or detailed
analysis of the recovered artifact collection will address this
question.

Archaeological excavations within the project area have
documented the presence of the collector/village subsistence and
settlement pattern during the Late Archaic. The artifact
assemblages from the Marthaller and Ritsch sites stand out as
clear examples of this pattern. Compared to assemblages from
sites dated to the Middle Archaic, there is marked increase in
the number and variety of chipped- and cobble-stone tools,
indicative of more intensive resource use and longer periods of
occupation. The presence of stone net- sinkers and fish remains
clearly demonstrates that the people of the Late Arachic were
exploiting the fisheries of the Rogue River. The presence of
heavy milling and grinding stones at these sites shows that
people were making investments in non-portable tools, in addition
to the time and energy that was spent constructing substantial
dwellings. Clearly, the sites of the Late Arachic era in the
project area had become villages in the sense that people spent
considerable time at these locales and returned to the same
locations year after year.

The archaeological evidence for the collector/village way-
of-life during this period in the form of housepits is confined
to the radiocarbon dated features at the Ritsch site at 1,400 BP .
This date is within the time frame of the Siskiyou pattern as
postulated by Connolly. Pettigrew and Lebow (1987) advanced a
theory that small housepit village sites may have originated much
earlier in the region, possibly as early as 5,000 BP . Although
no evidence from the excavated sites along the Rogue River
substantiates this hypothesis, the excavation of several rock
features at both Marial and Stratton Creek, dating to the Middle
Archaic, hint at the possibility. Never the less, the
preponderance of material evidence that defines the
collector/village pattern--f rom housepits to fishing implements--
appears in the archaeological record in the project area during
the Late Archaic.

55

The reasons for the change into the collector pattern during
the Late Archaic are not clear. Connolly has linked the change
from a long-lived, forager pattern to outsiders, either the
intrusion of the Athapaskan- speakers (the Gunther Pattern) , or to
influences from east of the Cascades. Pettigrew and Lebow argue
that the development was a natural progression over time within
the region, as evidenced by the historical changes in artifact
forms (especially projectile points) .

Perhaps the best explanation to fit the data that is
available at the present time, is the model presented by Winthrop
(1993) . Implicit in her model is the interaction of a variety
factors. These factors include: the changing environmental
conditions of the Middle Archaic that produced a more productive
and predictable food supply in the river valleys by the beginning
of the Late Archaic; a growing population that required an
intensive utilization of resources within more restricted
territories; the evolution of food processing and storage
techniques, and land management practices (e.g., field and forest
burning) ; and expanded contacts through trade with societies
outside of the area.

Were the Athapaskan- speakers responsible for the development
of the collector/village pattern along the Rogue River?
Certainly they brought with them ideas, technologies, language,
and customs that were different from those of the long-time
occupants of the area. Many of these traits were no doubt
borrowed or imitated by the resident population. But the
Athapaskans in turn must have been influenced by the people who
had a long history of adaptation to the landscape and resources
of southwestern Oregon. The current archaeological data points
to an evolutionary process that developed in place over the long
span of the Early and Middle Archaic. The pace of change
accelerated during the Late Archaic, partially in response to
expanded cultural contacts, and resulted in the way-of-life that
was practiced by both Athapaskan groups and Takelma speakers at
the time of the appearance of the first Euro-Americans in the
Rogue Valley.

The Ethnographic Era (250 BP to 150 BP)

The following is a discussion of archaeological sites along
the river that relate to the ethnographic record. This
discussion is based on the research of several anthropologists
earlier in this century (see Chapter III) who gleaned their
information from a handful of Native American informants,
primarily elderly women. These women were small children during
the time of the destruction of the native way-of-life in the
1850' s. Their recollections of the traditional Indian lifeways
(some of which they learned from their parents and grandparents) ,
along with their knowledge of village locations and other
placenames of importance, are the only first-hand Native American

56

accounts that we have. The way-of-life that they described
applies with certainty only to the relatively recent past. Thus,
the ethnographic-period discussion is limited to the one hundred
year time frame prior to 1850.

The ethnographic way-of-life may have differed from much of
the Late Archaic for several reasons. Prior to the full-scale
invasion of the Indian's ancestral territory beginning in 1850,
indirect Euro-American influences may have profoundly changed
native economic and social patterns. The effects of European
diseases on Native American cultures have been well documented
throughout much of North America back as far as the 16th century.
Although no direct evidence exists for such epidemics in
southwestern Oregon, they may well have occurred early on and
significantly altered the structure of the native societies. The
presence of Euro-Americans in large numbers- -to the north in the
Willamette Valley and to the south in California- -prior to the
gold rush in southwestern Oregon, also had indirect impacts on
the cultures of the region. Euro-American goods began to enter
the area through trade, and population shifts and subsequent
regrouping of Indian cultures occurred throughout the West.
These factors and others no doubt affected life in this period
for the native inhabitants who lived along the Rogue River.

Our knowledge of archaeological sites that date to this time
period along the river is derived from several sources: field
surveys conducted by the BLM during the past two decades,
information supplied by local artifact collectors and property
owners, and from the recently completed literature review and
field survey that were undertaken as a earlier phases of this
study.

The boundaries of BLM management include the Recreational
and the Wild sections of the Rogue River. The Recreational
section of the river extends from the mouth of the Applegate
River downriver to Grave Creek, below the town of Galice. This
portion of the river includes portions of the ethnographic
territories the of Penutian- speaking Lowland Takelma, and the
Athapascan- speaking Da-ku-be-te-de of the Applegate Valley, as
well as the Tal- tuc-tun-te-de people, whose primary village was
located at the mouth of Galice Creek. Downriver from Grave
Creek, the Athapascan- speaking Shasta Costa probably inhabited
sites within the Wild section of the river as far west as the
mouth of the Illinois River.

Taking the Recreational section of the river first, a few
ethnographically described village locations have been identified
with archaeological remains; several other identified sites may
also relate to this time period. The Marthaller and Ritsch sites
(discussed above) near the mouth of the Applegate River may be
the location of the ethnographically described village of the Da-
ku-be-te-de people. Two different Native American informants

57

placed the village of Da-ku-tee (also spelled La' kho-ve) at the
mouth of the Applegate River (Jacobs n.d. and Harrington
1981:347) . The Marthaller site seems like the most probable
corresponding location, due to its size and the density of
cultural material recovered during the test excavations. The
researchers of that excavation (Steele 1984) suggested that the
site may have been abandoned prior to contact with Euro-
Americans, due to the lack of recovery of any historic trade
goods. However, only a relatively small sample of the site was
excavated, and historic trade goods (e.g., metal, glass, or trade
beads) are seldom found in large quantities in sites in
southwestern Oregon. The Ritsch site may also date from this
era, although there again, no historic trade goods dating to the
contact era were recovered. The Ritsch site, however, had a
relatively low artifact density and was buried beneath layers of
sediment, which would argue against it being the major village
site of the Applegate Athapaskans . At this point, there is no
"smoking gun" to directly associate either site with the village
of Da-ku-tee, although both sites do fit the location description
of that village, and both were occupied into the Late Archaic.

A few miles downriver from the Marthaller site, the area
from Finley Bend to Pickett Creek, just below the Robertson
Bridge (a 6.5 mile stretch of the Rogue River), contains a number
of archaeological sites that have been documented or reported.
Although there is no direct correlation with ethnographically
described placenames or villages, the noted or described
artifacts and/or features at these locations point to a Late
Archaic occupation, probably by the Lowland Takelma. Two of the
sites were described by a local resident (Wallmann 1994 personal
communication) to have contained housepits. A site recorded by
the BLM in 1977 (Deich) contained deep midden soils, in addition
to notched net-sinkers and fire-cracked rock; two additional
sites recorded during the field inventory phase of this current
project also exhibited deep midden soils, fire-cracked rock, and
notched net-sinkers. Towards the downriver end of this section
of the river corridor, a complex of sites on a broad terrace was
reported. These included one site area from which obsidian
wealth-display blades, a pestle, a soap-stone pipe, and a stone
paddle-club were reportedly unearthed in the 1940' s. All of
these locations along this stretch of the river have been
disturbed by historic farming activities and/or modern
construction projects.

Moving a few miles downriver to near the confluence of Jump-
Off Joe Creek and the Rogue River, three sites identified in the
ethnographic accounts within Lowland Takelma territory were
recorded during the field inventory phase of the current project.
These included the villages of Hat' onkh near Jump-Off Joe Creek,
and Yawa-kh and Sumulkh near Hellgate Canyon. The Hat ' onkh site
was tentatively identified based on the correspondence between
the ethnographically described location and presence of midden

58

soils, fire-cracked rock, and chipped-stone debris spread across
a broad, high terrace along the Rogue River. A small, fresh-
water mussel shell midden may have been associated with this
site; it is located on the shore of the river directly opposite
Hat'onkh.

The villages of Yawa-kh and Sumulkh are more firmly
associated with the ethnographic record. Not only do the
locations of these sites conform to the Native informants'
descriptions, these sites also contain features and artifacts
dating to this era. At the Yawa-kh site, local relic collectors
report that in the 1950 's there were 16 intact housepits at the
site. The ethnographic accounts describe Yawa-kh as " ...a good
sized town." (A portion of the site was dug by artifact
collectors in the 1970's). Artifacts recovered at that time
included an iron adze blade, a rolled copper bead, and a piece of
clay pottery. This site was inspected during the course of the
current project. The excavations and backdirt piles from this
early vandalism are still evident, and fire-cracked rock and
chipped-stone flakes were noted eroding from the sides of the old
excavation units.

The site of Sumulkh, on the opposite shore of the river from
Yawa-kh, also contained a number of housepits that were intact in
the 1950' s. In the last few decades highway and residential
construction has destroyed much of this site. BLM site
information from the 1970 's reported the presence of projectile
points, fire-cracked rock, notched net-sinkers, and historic
glass trade beads. During the field inventory phase of this
project, the site was revisited; thick midden soils, fire-cracked
rock, a basal-notched projectile point, a notched net-sinker, and
the possible location of one housepit were noted. Although both
the Yawa-kh and Sumulkh sites have been severely damaged, there
still remains a potential for future scientific investigation of
the intact portions of these sites.

The last site in the Recreational section of the river
corridor tied to the ethnographic period is the village of
Talda' cdan (or Talustun) near Galice Creek. The village,
according to the ethnographic accounts, was located on both sides
of the Rogue River, and was the principal village of the
Athapaskan- speakers who occupied the Galice Creek drainage. The
site was recorded by the BLM (Deich 1977; Winthrop 1993a) .
Artifacts noted at the site included waste flakes from stone tool
manufacture, fire-cracked rock, animal bone fragments, and fresh-
water mussel shells. Unfortunately, large portions of this site
have also been vandalized.

The ethnographic data and the archaeological information for
the Wild section of the Rogue River between Grave Creek and
Marial is limited. There is a Lowland Takelma reference for what
may be Rainie Falls, Taktkamaykh . Informants referred to it as

59

"a portage of canoes and big waterfall way down Rogue River
(Harrington 1981:509) ." This would have been at or near the
western boundary of Takelma territory. Although Dorsey (1890)
provides a list of Shasta Costa placenames on the Rogue upriver
from the mouth of the Illinois River, their relation to specific
locales within the Rogue River Canyon are undetermined.
Archaeologically, a few sites have either been reported by relic
collectors or noted by BLM personnel between Grave Creek and
Marial . The lack of suitable landforms (large terraces above
high water levels) , combined with the difficulty of access within
the canyon may have precluded intensive settlement . There are
reports of archaeological remains at several of the larger bars
and terraces near creek mouths; however, no dates have been
ascribed to these sites.

There are a number of sites along the river within the
project boundaries that date to at least the Late Archaic. Of
that number, a few can be directly associated with the time
period immediately preceding the destruction of the traditional
way-of-life in the 1850' s. The sites of Yawa-kh, Sumulkh, and
Talda' cdan all have direct, specific ethnographic references; in
addition, historic trade goods have been reportedly recovered
from two of these sites. Less well documented evidence, but none
the less suggestive, exists for the association of the village of
Da-ku-tee with the Marthaller site, or possibly the Ritsch site;
and the village of Hat ' onkh with the site recorded near Jump-Off
Joe Creek. The other sites, either documented during the field
inventory or reported by collectors, cannot be dated to such a
narrow time frame.

The locations of these Late Archaic, ethnographic-era sites
provides us with some insights into land use patterns that help
to predict the location of additional sites along the river. One
interesting pattern appears to be the pairing of sites on
opposite sides on the river. This pattern occurs at the mouth of
the Applegate (the Marthaller and Ritsch sites) , and at four, and
perhaps five other locations along the Rogue between the
Applegate River and Galice Creek. Most of these sites are also
located alongside a tributary creek of the river. It is the
sediments (rocks and gravel) that are discharged from tributary
creeks during flood episodes that form the rapids and riffles of
the Rogue that in turn create the pools that provide excellent
fishing potential. Two of these locations, at the mouth of
Galice Creek and the villages of Sumulkh and Yawa-kh. appear to
have been occupied concurrently during the ethnographic era.

Intuitively, it seems that the area between the confluence
of the Applegate and Rogue Rivers and Galice Creek was
intensively occupied during the Late Archaic and/or ethnographic
eras; thus, populations may have been relatively high. However,
solid evidence for this hypothesis is still lacking. We have yet
to firmly date most of these sites to a narrow time frame.

60

Perhaps the sites were all occupied during the 100 years of the
ethnographic era. Even if that is true, we do not know if use
occurred each year at every site, or whether sites in different
locations were utilized on a sporadic or rotating basis depending
upon resource availability, environmental conditions, or cultural
traditions .

Archaeologically, population estimates based on the number
of housepit features reported at various sites can, at times, be
misleading. An example from the upper Applegate River Valley is
instructive in this regard. At site 35JA42, five housepit
depressions were discovered during the course of an
archaeological inventory. After excavation, it was determined
that, in fact, the village was only one house in size. It seems
that the occupants of the site rebuilt a single new house each
year, abandoning the previous year's dwelling (Brauner 1983).
Although the surface indications pointed to a small village site,
the reality was a single family residence. Thus, the size of the
resident population and the intensity of the settlement within
the project area based on presumed village size must await a more
exhaustive examination of the archaeological data.

Due to the rapid dislocation of the Indian population after
Euro-American contact, estimates of the indigenous population of
the region vary widely. However, Beckham (1971:9) puts the total
population of Native Americans living in southwestern Oregon at
the time of Euro-American contact at nearly 10,000. If only a
fraction of this number resided in the project area, it is likely
that several thousand native people lived along the middle
stretch of the Rogue River. Although by today's standards this
appears to be a relatively low population density, in 1870,
twenty years after the beginning of Euro-American settlement of
the region, Josephine County's official population stood at only
1,204 (U.S. Bureau of Census 1870). Native populations, prior to
contact with Euro-Americans, may well have rivaled or exceeded
those of the early historic period.

War and Diaspora (1850 AD to the Present)

Early conflicts between the Native Americans and the
occasional fur trappers and traders who ventured through the
region in the 1820' s and 1830' s set the tone for future relations
between the two cultures. The discovery of gold in California in
1848 prompted a flow of would-be miners from the Willamette
Valley to head south through the Rogue Valley on their way to the
southern gold fields. This intrusion through Indian territory
was only the beginning. By 1850 settlers were beginning to enter
the Rogue River region in increasing numbers to take up land,
spurred by the passage of the Donation Land Act that (amended)
granted 160 acres (320 for married couples) to homesteaders.
With the discovery of gold on Jackson Creek in late 1851, the
trickle of immigrants turned into a torrent. Within the space of

61

a few months, miners had stake claims throughout southwestern
Oregon, and boom towns had sprung up along the rivers and creeks.
Increasingly, the Indians found themselves dispossessed from
their ancestral lands.

Conflicts between the cultures escalated until warfare broke
out in the Rogue Valley during the summer of 1853. As noted by
the historian Steven Beckham (1971) the reasons for the Indians
resentment of the miners and settlers were well founded:

Above all else, the Indians in the region had been reduced
to starvation and to surviving virtually as refugees in
their own homeland. The miners' debris had ruined the fish
runs. The settlers' hogs had eaten the acorns and the
cattle had cropped off the camas . The pioneers had erected
laboriously-made, split-rail fences and would not let the
Indians burn the fields and hillsides as they had done for
ages to produce new seed crops or keep down the brush for
good hunting. Further, the whites with their modern weapons
had taken a heavy toll on the deer and elk in the region.
Repeatedly the Indians had been driven from their old
villages as farmers moved in to file for lands under the
Donation Land Act of 1850, or as miners decided to overturn
the river bars in search of gold.

The hostilities of 1853 were brutal but brief. The Indians
of the interior Rogue Valley signed a treaty ceding most of the
Rogue Valley in return for a reservation near Table Rock. That
same year over on the coast, gold was discovered in the beach
sands along the southern Oregon coast . The rush of miners to the
coast had the same effect on the coastal Indian groups; violence
and warfare ensued.

Life on the Table Rock Reservation during 1854-55 had
devastating consequences for the native population. Lack of
food, a severe winter, and diseases ravaged the people. In
October of 1855, when a group of Takelma returned to their old
village at the mouth of Little Butte Creek on the upper Rogue
River, they were attacked by a group of volunteer militia. Eight
Takelma men and 15 women and children died in the attack. This
event led to breakdown of any pretense of coexistence between the
cultures. The Indians at the Table Rock Reservation, sensing
that survival was no longer possible in their homeland, headed
for the security of the Rogue River Canyon and the coastal
mountains to the west. Their flight from the reservation was not
without revenge; they attacked miners and settlers along the
Rogue River from Gold Hill to Galice Creek. The final battles of
the war now moved into the country surrounding the middle course
of the Rogue River, a rugged terrain of steep slopes and narrow
canyons .

62

Within the project boundaries, the attack by the westward
fleeing Indians on the mining community of Galice in late 1855
has been well documented. On October 17, Indians attacked the
mining encampment at the mouth of Galice Creek, a strategic point
for access to the Rogue River Canyon. The forty or so miners
and packers at Galice defended their settlement during the day
long battle, but suffered a number of casualties, including the
death of four of the defenders (Beckham 1971) . The Indians
withdrew to fight another day.

Moving west to the rugged mountainous terrain north of the
Rogue River, the Rogue River Indian groups established camp on a
high ridge. Discovered by the pursuing volunteer militia and
regular army troops during the last days of October 1855, battle
ensued. The Indians held the high ground and inflicted heavy
casualties on the enthusiastic, but poorly trained troops. The
Battle of Hungry Hill, as it was known by the defeated, was a
major victory for the Indian forces. The final military campaign
of 1855 against the beleaguered Indian forces commenced in the
later part of November. Regular troops and volunteer militia
advanced from Grave Creek, down the Rogue River Canyon and
eventually encountered the Indians below Whiskey Creek. Nearly
150 Indian men and their families were camped on the south side
of the river at Black Bar. The soldiers attempted to launch a
surprise attack on the encampment from the north shore, but were
given away by the noise of their axes as they felled trees to
build rafts. The Indians opened fire on the exposed positions of
the volunteers, defeating their pursuers for the second time in a
month. With the onset of winter weather, hostilities within the
canyon came to an end.

In the following spring of 1856 the final engagements of the
war were fought. Within the project area, Indian forces from the
Rogue Valley suffered their first defeat since leaving the Table
Rock Reservation. A combined force of nearly two hundred
volunteers reached Ditch Creek on the north shore of the Rogue
and fired into the Indian camp on the opposite shore. Perhaps as
many as thirty warriors were killed during the all-day assault at
what today is named Battle Bar. This victory by the militia,
combined with the hardships suffered during the long winter,
forced the surviving Indians to capitulate. One final desperate
battle was fought at Big Bend on the lower Rogue River, when
surrendering Indians were attacked by volunteer forces. The end
of May, 1856 saw the cessation of formal hostilities. Throughout
that the summer and the following winter, "licensed" Indian
hunters killed and captured scattered Indian survivors throughout
the region (Beckham 1971) .

Following surrender, those Indians who had survived war,
privation, and disease were forced to face another trial, removal
to a reservation. This process had begun earlier with Takelma
and Shastan people who had not fled the reservation at Table

63

Rock. In the winter of 1856, 400 Indians from Table Rock were
marched 200 miles overland to the newly established reservation
of Grande Ronde in Yamhill County in the Willamette Valley.
There, they resided with Indians from the Willamette Valley and
the lower Columbia River (Beckham 1978) . Those bands that
surrendered in the spring of 1856 at Big Bend, including
survivors from Takelma and Shasta Costa villages, were loaded
onto a steamer at Port Orford and transported, via the Columbia
and Willamette Rivers and then overland, to their new home, the
Siletz Reservation near the Oregon coast. One group of survivors
from the Applegate Valley (35 men, 90 women, and 90 children) led
by their leader "John", made the trip to Siletz in a different
fashion; they were forced to march 125 miles up the Oregon coast
(Beckham 1971) .

By all accounts, the early years at these new reservations
were a continuation of the degradations and deprivations that had
ravaged these traditional societies for the preceding decade.
Forced to live in an unfamiliar environment among people with
different backgrounds and traditions, the Indians of southwestern
Oregon lived and died in despair. Disease ravaged the
populations at both reservation. At Siletz the population fell
from 2,026 in 1856 to 483 in 1900. The story was similar at
Grande Ronde; 826 people in 1857 and only 398 surviving in 1910
(Beckham 1978) . Most of these deaths occurred shortly after the
survivors of the Rogue River hostilities arrived at the
reservations, as noted by the Agent (Metcalf) at the Siletz
Reservation in 1858:

According to the census taken of those people [Rogue River
Indians] twelve months ago, they numbered five hundred and
ninety, out of this number two hundred and five have died;
-■ thirty-five have returned to Grande Ronde, and three hundred
and fifty remain, many of whom are sick. . . (Beckham 1978) .

The Euro-American ideals of "civilization" were forced on
the survivors. Christianity, speaking English, and farming were
the goals of the reservation system. Children were separated
from their parents and placed in boarding or day schools to keep
them from the being influenced by the elders (Beckham 1978) .
Gradually the native languages, oral traditions, customs, and
knowledge of the traditional ways-of-life were diminished with
each generation.

The twentieth century has been a long struggle from the edge
of extinction to the promise of a new era for descendants of the
original inhabitants of the Rogue River. The members of the
Siletz and Grande Ronde fought back from the official termination
of their tribal status in the 1950' s to reclaim a portion of
lands allotted to them in the nineteenth century. Today, tribal

64

governments increasingly control their political and economic
future, and they have developed programs to revive cultural
traditions and to protect sites and landscapes of sacred value.
Although the smoke of sawmills and smudge pots has replaced the
haze of village campfires and meadow burnings, the memory of the
people for their homeland along the Rogue has not diminished.
Even while a new culture occupies what was once their land, the
old ties have not been broken. It is this long- lasting
identification with the land of their ancestors that has provided
the Indians of the Rogue River Valley with the ability to cope
with the challenges of the changing modern world.

65

PART II: THE HISTORIC ERA

Introduction

Part I of People and the River reviews nearly 8,000 years of
history of the Rogue between the Applegate River and Mule Creek.
During these millennia Native Americans successfully sustained
life in villages or at seasonal camp sites along the stream.
Part II traces Euro-American occupation of the canyon for the
past one-hundred and sixty years. Although it covers a much
shorter period, this overview adds a distinct new chapter to the
Rogue's story. The years between 1850 and the present are filled
with human ventures -- exploration, gold mining, war,
agriculture, settlement, recreation, and governmental management.
These endeavors emerge at various times and places, establishing
repetitive themes which weave through this segment of the river's
history.

In Part II, which covers the years 1827 to the present, the
stream's middle course is divided into two parts. Chapter Six
reviews the section of river which extends between the mouth of
the Applegate River and Grave Creek, an area more accessible, and
hence, more settled, than lower reaches of the canyon. Chapter
Seven traces the history of the rugged and isolated stretch of
river between Grave Creek and the mouth of Mule Creek. In this
reach, human occupancy has generally been more sparse. Chapter
Eight surveys the entire study area in the years since World War
II.

The Rogue River's long story is about geology, floods, flora
and fauna. It is also about people who have sought its wealth,
isolation and beauty for at least 8,000 years -- the canyon
residents for whom life has centered along the stream. Along the
river's banks human's have sought shelter, warmth, and food. Its
bars and terraces have been used over and over; its fish have
provided nourishment for food or sport; its rock walls have
shielded people who sought seclusion or protection. This modern
chapter of the Rogue's story begins early in the nineteenth
century when Euro-Americans arrived in southwest Oregon for the
first time.

ig

Figure 6. Map of the Rogue River from the Applegate River to
Hellgate Canyon

67

VI. APPLEGATE RIVER TO GRAVE CREEK

Exploration

In the spring of 1827 Indians gathered anxiously at their
camp a short distance downstream from the mouth of the Applegate
River. They used the site regularly in the summer and fall while
dip-netting salmon and processing game and cooking camas . This
spring the fire hearths, grinding slabs, pestles, and scrapers
again awaited summer use. A day earlier, this small band had
been stunned to see a large group of strangers ford the Rogue
about three miles to the east.^ Who were these sallow- skinned
men, women and children who now invaded the valley?

The newcomers' shouts had shattered the silence. Some of
them rode horses -- animals unfamiliar to the native
inhabitants. Their clothing of fur and bright-colored fabric
astonished the observers. The Indians heard loud explosions and
saw deer fall at the sound. Most disconcerting of all were the
leather tents pitched near the stream. These intruders behaved
as if they intended to stay.

Peter Skene Ogden, Hudson's Bay Company employee and leader
of the group, was equally wary of the Indians. Although he was
nervous about taking too much time in the region, his assignment
to trap beaver meant that the streams must be explored.
Disconcerted by the unfamiliar topography, Ogden hoped to learn
of the rivers and trails from the Indians. Soon after fording the
Rogue he wrote :

. . . Indians have all scattered themselves and we cannot
procure one to Guide us this places us in rather an
awkward situation and I feel rather at a loss how to
act... I sent three men down the main Stream in quest
of a Guide and they succeeded in procuring one, this
certainly should the fellow not escape be of a great
advantage to us he appears rather alarmed and wild we
must endeavour to tame him. . . (Davies 1961:91)

Ogden crossed the Rogue River on March 9th, 1827
approximately one mile east of its confluence with the Applegate
River (LaLande 1987:65). Some of his scouts briefly separated
from the main party to view the mouth of the Applegate River
which lay near the Indian's camp, now known by archaeologists as
the Marthaller site.

^ Ogden' s party consisted of nearly two dozen trappers, their
Indian wives and children, totalling nearly one hundred people as
well as large numbers of pack horses.

68

Following Ogden's visit, Hudson's Bay Company trappers
seasonally traversed the Rogue Country in search of beaver. The
company policy focused on exhausting fur resources throughout the
northwest in order to discourage American trappers and ensure
British control of the region. While the number of beaver
removed from the Rogue is unknown, frequent trapping undoubtedly
reduced the number of fur-bearing mammals along the stream.
Ogden reflected on the effects of trapping while in the Rogue
country (Davies 1961:91)

We have this day [March 11, 1827] , eleven Beaver it is
almost a sin to see the number of Small Beaver we
destroy and to no purpose some of the Females taken
have no less than five young and on an average three
each in a month hence their young would be sent in to
the world this is the affect of Traps it spares neither
Male or Female almost equal to death all are distroyed.

For thousands of years the Rogue corridor had sustained life
for native peoples who found abundant fish, game and edible
vegetation along the river and in the nearby hills. Now, traps
gripping broken bodied mammals were pulled from the stream.

Ogden's journey was the first known fur-trading expedition
through the Indians' territory and the first effort to extract
the natural resources of the area. As the years passed, the
relentless trapping had a distinct economic and environmental
impact, initiating a new era which directly conflicted with the
subsistence living pattern developed by native peoples.

During the twenty-five years between Ogden's 1827
expedition and a local gold discovery in 1851, the native
inhabitants saw many strangers traverse the ■ Rogue country.
Trappers, government explorers, entrepreneurs, and gold miners
regularly used the path once known only to the Indians .
Gradually this ancient north- south route was developed into the
Oregon-California Trail.

In 1829 Ogden's fellow Hudson's Bay Company trapper,
Alexander Roderick McLeod, crossed into the Rogue drainage,
probably reaching the mouth of the Applegate River (Beckham
1971:32). During the 1830' s and 1840' s others passed through
Southern Oregon, trapping, hunting, and occasionally clashing
with inhabitants. Steven Dow Beckham (Beckham 1971:32) noted
that by 1833 "Perhaps as many as 250 had passed over it and
through the Rogue Valley. . . "

In 1833 fur trapper Michel LaFrambroise descended the Oregon
Coast to California, and later led his men north through the
Rogue River Valley. John Work, another Hudson Bay Company
brigade leader, left the Sacramento Valley in mid- summer, 1833
for Fort Vancouver and camped near the Rogue River, leaving the

69

area on October 3, 1833. The same year American trapper Ewing
Young led a party through the Rogue Valley. Young returned to
Oregon over the same route in 1834 and in 1837 led a large cattle
drive over the trail from California to the Willamette Valley.

In September 1841, members of the U.S. Navy Exploring
Expedition, led by George F. Emmons, traveled through the Rogue
River Valley as they explored the land route between the Columbia
River and San Francisco Bay. Expedition members recorded their
impressions of the area as they passed through. As others had
done previously, the expedition crossed the Rogue a short
distance east of the mouth of the Applegate River. On September
25th, Titian Ramsay Peale, artist and naturalist, wrote
{Poesch:1962:191) :

Reached and crossed "rogues River" before night,
pitching our tents on its South bank. Some Indians
approached in canoes, but were not suffered to enter
the camp. The river was at the camp about 90 yards
Wide, and three feet deep with a gentle current, and an
even, gravelly bottom.

American trapper James Clyman led a large party across the
Rogue River on his way to California in 1845 -- the last
documented fur brigade through the area before the establishment
of the Applegate Trail. In June, 1846 Jesse and Lindsay
Applegate, Levi Scott, and others forged a wagon route along the
Oregon-California Trail before turning east through the Cascade
mountains toward Klamath Lake. In the late fall of 1846 Jesse
Applegate returned over the trail leading one-hundred wagons back
over the southern route to the Willamette Valley. Crossing the
Rogue River at the well -used ford they headed north to Grave
Creek where heavy rains, illness and exhaustion overcame the
travelers. After suffering through a wet winter, surviving
members of the party reached the Willamette Valley.

Prior to contact with Euro-Americans in the early 19th
century the Indian people survived for thousands of years in a
successful balance with nature. The fur-traders instilled fear
and distraction and within a few decades the native culture
disintegrated as the seasonal rhythms were broken forever.

Mining

Shortly after the last beaver were pulled from the Rogue, a
new source of wealth was found in the water. The 1848 gold
discovery in California brought Willamette Valley farmers- turned-
miners south over the California-Oregon Trail into the Rogue
country. They traveled any time of year and they came in hordes.
Prospectors explored the Rogue River and its tributaries on their
way south and early in 1852, gold was discovered near the present

70

town of Jacksonville. Soon miners swarmed over streams
throughout the region.

Like the trappers who took beaver, or "soft gold" from the
water, the miners wrested wealth from the river. Mining
intensified the pattern of resource extraction begun by the fur-
traders. At places where Indians had fished for salmon, men now
demolished the streambanks in their search for gold. Permanent
Euro-American occupancy of the Rogue country abruptly took hold
as gold miners broke new trails through the hills and along the
river's edge. The sounds of metal on rock signaled new residents
in the canyon.

While the earliest miners used picks, shovels and pans along
the Rogue, more advanced equipment soon followed. Rockers --
wooden cradles with one open end -- were manipulated by hand and
processed several cubic yards of dirt in a day. Sluice boxes,
which were troughs usually ten to twelve feet in length, were
placed end-to-end with a drop between each section. Riffles on
the bottom caught the gold. (Follansbee; Pollock 1978:170) In
operations of any size, several men were needed for digging
ditches, building flumes and operating the equipment.

Although mining yielded generous returns during the early
years the richest prospects were soon exhausted. New
technologies of hydraulic and dredge mining eventually boosted
placer mining productivity in the area. Remaining evidence of

Galice, ca. 1910

Siskiyou National Forest Photo

71

extensive hydraulic mining can be seen in several locations along
the Rogue in the form of tailings, rock walls and ditches.
Tailing piles are especially apparent near Galice.

Established early in the 1850' s, Galice was named for Louis
Galice, a French placer miner who first prospected the creek
which bears his name. Galice Creek and its tributary gulches
were intensively mined and the small community quickly became an
important center for the region's mining population. Here the
Galice Store, succeeded by the Barlow Store, sold food and goods
to miners and packers .

A devastating flood in 1861 caused great damage at Galice
and to the rest of the river canyon. Deep snow, followed by warm
rain, flooded the bench land along the river. One traveler
described the Rogue in flood (Philip Eastwick in Atwood
1978:233) :

. . .All the rapids and falls of the river are lost sight
of, and the river assumes the character of a boiling,
surging mountain torrent filled with strong eddies and
whirlpools, and carrying down with it immense
quantities of drift wood to the ocean."

The flood destroyed placer mining operations along the river
as it washed out dams, ditches, and sluices. While damage to
buildings and equipment was serious, the ravaged streambanks and
altered creek channels provided fresh locations for miners to
work. A newspaper account focused on the destruction:

Of the eight houses at the mouth of Galice Creek, but
one is left standing. All the dams on that stream have
been swept away. At Skull Bar, eight houses have
disappeared, and but two remain on the Bar, one of which
is the store of J.V.R. Witt; his goods are undamaged.
At this place the water, Sunday noon, was 45 to 50 feet
above low water mark. Five houses floated by within an
hour {Oregon Sentinel December 21 1861) .

The abrupt influx of miners into the Rogue country
devastated local Indian bands. Miners' ruthless treatment of the
native people and their steady assault on the environment brought
chaos to the canyon. Streambank fishing sites were destroyed by
mining and upland camps disappeared when trees were stripped from
the hills. The rapid loss of their fishing and hunting grounds
brought bitter clashes between the Indians and United States
government and volunteer forces between 1851 and 1856. The major
conflicts which occurred along the Rogue during these years are
described in Part One of People and the River. By spring, 1856
the Indians were defeated. The straggling few who survived the
doomed struggle were taken to a place they did not know. They no

72

longer cut willow and alder for baskets and their stone
implements lay abandoned at the camps beside the river.

73

Figure 7. Map of the Rogue River from Hellgate Canyon to Grave
Creek

14

Forty years following the outrage which forced Indians from
their homeland, another inequity occurred -- one more dismal
injustice for people who had forfeited everything. Indian Mary
Park remains, ironically, as both the location where the episode
occurred, and as the site which commemorates the individual who
suffered the loss.

For many years prior to the construction of Hellgate Bridge
in 1913, a ferry operated at the present site of Indian Mary
Park. There, from approximately 1860 until his death in 1886,
"Umpqua Joe" farmed, ran the ferry, and lived nearby. According
to local legend, Umpqua Joe, who had warned Gal ice miners of an
impending Indian attack, was allowed to remain with his wife and
small daughter on the land, after other local Indians were
removed. Umpqua Joe's continued residence there was one of the
few instances where native people were allowed to continue their
tradition.

Following her father's death in 1886, his daughter, Mary,
who had been born on the land about 1856, received a grant with
homestead rights from the federal government. She continued to
run the ferry until she moved to Grants Pass in 1894, when she
leased it to William A. Massie. Following Mary's death, local
authorities determined that she had no title to the land and her
heirs were not recognized. This action constituted the last
instance in which an Indian land holding passed to Euro-American
hands .

The land which Umpqua Joe's family had occupied for fifty
years was eventually developed as a park by Josephine County.
Galice Store owner Frank Carpenter's home was purchased by the
county and moved to Indian Mary Park to serve as the caretaker's
house .

After Indians were removed from the Rogue country in 1856
the region was fully opened to miners, farmers and entrepreneurs.
Mining developed at a steady pace along the Rogue as important
gold discoveries occurred. In late 1874 the Mammoth and Yank
Ledges (also known as the Big Yank Lode) were discovered with
placer ground extending across the bed of the Rogue River. In
less than a month 200 claims were taken on these immense veins
and the roads were choked with teams heading toward the new
workings .

Quartzville, a new town at the mines, was surveyed into lots
and Yankville was established a mile above the site. (Both
communities were situated in the vicinity of Rand and Almeda
Park) (Walling 1883) . When the ledge was discovered and
competition for ground increased, miners on Galice Creek met at
Sanders Store to regulate quartz mining in the district. In
forming the Yank mining district the committee composed a
preamble :

75

We, the committee appointed to draft resolutions beg
leave to present to this meeting the following
resolutions, concerning the quartz ledge discovered
some three miles below the mouth of Galice Creek on the
8th day of August, A.D., 1874, by "Yank" McNair and Co.
Resolved, First, this quartz ledge shall be known as
the Yank ledge (Grants Pass Bulletin January 29, 1937) .

Although official formation of the Almeda Mining Company did
not occur until the end of the century, early exploration of the
mine's location occurred with discovery of the Yank Ledge.
Eventually the site would become one of the most extensively
developed mines in southwestern Oregon, yielding valuable
quantities of copper, gold, silver, lead and zinc (Libby 1967) .

Hydraulic mining continued in several locations along the
Rogue in the early twentieth century as developing technology
made it profitable to mine the lower grade placer deposits.
The Old Channel Hydraulic Mining Company worked a terrace for two
miles along the Rogue River at Galice Creek. Nearby, the Galice
Consolidated Mines Company owned much of the creek's placer
ground. (Beckham 1978: 107)

Other rich mining sites were found in bench gravels at
various places on the Rogue River in the vicinity of Pickett
Creek, Hellgate Canyon, and Galice. The Flanagan Mine, the
Robert Dean Placer Mine and the Hellgate Placer Mine exemplify
extensive hydraulic mining operations near these locations.

The Flanagan Mine, located on the west bank of the Rogue
River about one mile upriver from Robertson Bridge consists of
about 200 acres (Oregon Metal Mines Handbook 1952) . This
extensive mine site retains several extant resources including a
number of mining faces, a large area of tailings and the partial
remains of water ditches.

Dr. William Flanagan, prominent Grants Pass physician, came
to Grants Pass in 1884 and opened a medical practice which he
pursued until his death on February 18, 1930. In addition to
holding interest in this extensive hydraulic mining operation,
William Flanagan served as mayor of Grants Pass and as Josephine
County Coroner (Atwood 1983) . In 1937 his widow sold a portion
of the mine and production most likely ceased by the end of the
decade .

Another long-active mine site is located several miles
downstream from the Flanagan Mine on the right bank of the river
near Rand. The Robert Dean Placer Mine was established on
September 8, 1900 by Lydia Dean and H.A. Corliss. By 1910 the
claim included one and three-quarters miles of ditch, a
reservoir, hydraulic giant and pipe, two cabins and six placer
excavations. The mine was run by Dean family members until 1940.

76

That year there were four and one-half miles of ditch, 2000 feet
of eleven to twenty inch pipe, and four hydraulic giants in
operation.

Hydraulic Mining, ca. 1900

Four structures at the Robert Dean Placer Mine, including a
two- story house and a log building, were dismantled and burned by
the Bureau of Land Management in the late 1970' s. A recent field
survey suggests that subsequent mining activity has obliterated
many of the older features. Extant improvements at the placer
mine include water ditches, a reservoir, cement headgate and
traces of former residences.

Gold-bearing lode mines were discovered at several places
near Galice and at sites downriver. Stephen Dow Beckham
(1978:107) notes that lode mining, "included fairly extensive
milling and processing activities at the Gold Bug, Golden Wedge,
Sugar Pine, Almeda, Oriele, and Bunker Hill Mines."

Both large-scale hydraulic and lode mining operations
incurred high equipment costs and demanded an increased labor
force. These additional requirements usually meant that operators
sought corporate backing. After 1900 several mining promoters

77

appeared in the Rogue River country and formed stock companies to
mine on a very much larger scale than had ever before been
attempted (Billings 1974) . These large lode mines had important
economic consequences for the Rogue country. People found
immediate work as carpenters, miners and packers assisting on
mining and construction projects.

The Almeda Mine was one of the most extensively developed
mines in southwestern Oregon, with large quantities of copper,
gold, silver, lead and zinc extracted from nearly one and one
half miles of underground workings (Libby 1967:6) . During its
principal period of operation between 1905 and 1916, production
at the Almeda Mine was valued at more than a hundred thousand
dollars {Oregon Metal Mines Handbook 1952: 17-18). Financial
difficulties forced closure of the mine in 1917.

Almeda Mine, ca. 1910

The Yank Ledge, mentioned earlier, and also known as the Big
Yank Lode, had been discovered by prospectors in 1874. While the
site had raised intense interest, a lack of sophisticated

78

equipment slowed exploitation. Late in the century the Almeda
Mining Company was formed by an Ashland, Oregon group. In
November, 1903, the Pacific Miner publicized the Almeda Mine in
enthusiastic terms:

[In the Galice District] "exists a mineral ledge of
gold, silver and copper that for size has no equal in
the State -it is called the Big Yank Mine... about
three miles down Rogue River from Galice Creek. . .Today
the Almeda Mining Company, a corporation founded by Mr.
Wickham and friends, with a capitalization of
$1,000,000 have a mine second to none in Southern
Oregon .

In July, 1906 four lode claims and two placer claims
comprising the Almeda Mining Company holdings were conveyed to
the Almeda Consolidated Mining Co. The lode claims included the
Monte Cristo, the Bonanza Lode, the Live Yankee, and the Yankee
Doodle, all claimed in 1899 and 1900. At the times of the
conveyance, O.M. Crouch was president, R.C. Kinney and John F.
Wickham were officers. Wickham' s son. Perry R. Wickham, served as
superintendent from 1905 to 1911; P.H. Holdsworth succeeded
Wickham as superintendent in 1911 (Libby 1967:6). A promotional
brochure published in 1909 described the Almeda Mines' facilities
at that time:

Two hundred acres of gold placer ground, with water
rights, ditches, and hydraulic equipment, also roads,
buildings, mine equipment, including tools, tracks,
cars, air compressor and drills, two gasoline and two
steam engines, hoist, sawmill, shingle mill and planer,
logging equipment, teams and wagons, assay office, and
100 ton smelter.

A log bridge was constructed across the river in 1909-1910,
and construction materials and equipment were hauled to it from
the railroad station at Merlin (Keyte in Hill 1976:63) Many of
the mine's employees resided at Rand which was situated a short
distance away.

Financial difficulties beset the Almeda Mine and on August
21, 1913 the mine went into the hands of a receiver on petition
of the State Corporation Commissioner and by order of the
Josephine County Court. On May 2, 1916 newspapers published a
notice of a Receiver's Sale, proceeds from which were to pay off
indebtedness of about $259,000 (Libby 1967:7) The last shipment
from the mine was made on July 28, 1917. The Almeda Post Office
which had opened on February 12, 1912 closed in July, 1920.
Termination of work at the Almeda Mine had greatly reduced the
area's population (McArthur 1992:14)

79

The Almeda Mine remained dormant between 1917 and 1937
despite several attempts to reorganize the operation. In the
late 1930 's the mine was briefly reactivated (McArthur 1992:14).
Shipments were made until 1942 when War Production Board Order L-
208 ended the sale of gold and the mine closed down. Today the
large earth scar which remains at the Almeda Mine is easily
viewed from the Rogue or from the Merlin-Galice Road.

Settlement

On the heels of the miners came the farmers. In a place
where landscape had long been defined by creeks, swales and
hills, newcomers now claimed the level, fertile lands between
the mouth of the Applegate River and Pickett Creek for private
ownership -- squaring it up, measuring lines on the ground and
laying out fields and fences to mark property boundaries.

''i,//i>*';'!v^''

i'^\

^fl. -A.- ^''^'^

A^'h

\V'

■^

*o.i^.

T^^' y'M

Zj,X.f/

^zy:^

-■-.->©/<«'

i3.%0

^<^.

.'^'^n

'^'"'^W^''"

/

.J

-"^^0-

nV.,

> //i^'

'4 '^T*^^^-''/ / //^•*^"^'r^

TV

General Land Office Map, 1854 (T 36 South, R 6 West, W.M)

80

The Donation Land Claim Act, passed by Congress in 1850,
offered potential residents free, farmable land. In its amended
form, the act promised 320 acres of land to a married couple and
160 acres to a single person if they would reside on the property
and construct improvements within a proscribed time. Between 1852
and 1855 at least fifty-five individuals staked donation land
claims in Josephine County.

Several settlers selected land along the Rogue River between
the mouth of the Applegate River and Pickett Creek (General Land
Office Survey 1854-1855) . The rich soil there promised good
crops with convenient access to water for irrigation. In August,

1852 James N. Vannoy staked a claim just east of the mouth of the
Applegate River, acquiring the land where the river ford was
located. Margaret Dimmick, James T. and Thomas Frizzell, and
Israel Elliott settled nearby in 1853 and 1854.

Others chose land downriver near the present site of
Robertson Bridge. Timolean Love, Moses Pasley and John Butcher,
for whom Dutcher Creek is named, took claims in 1855. George P.
Miller reserved his donation land claim in June of the same year
(Hill 1976:28-29) The newcomers built houses, barns, granaries,
smokehouses and irrigation ditches. In the broad fields and
pastures they raised grain crops, corn, apples, pears, cattle,
and hogs .

High water occurred frequently on the Rogue during the early
years of Euro-American settlement. Accumulated snow in December,
1852, subsequent warming weather, and heavy rains in January,

1853 combined to cause severe flooding in the Bear Creek Valley
and along the Rogue in what is now Josephine County. In March,
1858 the Rogue River and other southwestern Oregon streams
swelled following severe rains, which "caused a flood in almost
all streams" (Crescent City Herald March 24, 1858) .

Severe flooding in December, 1861 inundated fields along the
Rogue River plain west of present Grants Pass and destroyed
improvements and crops in the agricultural section from the mouth
of the Applegate to the mouth of Jumpoff Joe Creek. Along the
Rogue River below Hellgate the rushing waters washed out cabins
and sent mining equipment splintering over boulders.

"FLOODS -- GREAT DESTRUCTION OF PROPERTY

Mr. Vannoy lost his ferry, fifty head of sheep, and a
lot of other stock, with his house furniture, hay and
oats... Mr. R.S. Belknap carried his family to high
land, from his house on horses. His valuable orchard is
almost entirely destroyed. His loss in fence, hay and
grain is heavy. Miller and Gibson have also sustained
heavy losses in hay and grain. At the mouth of Limpy
Creek, the Dutcher and Shan Evans places are covered

81

with sediment from six inches to two feet in depth.
About thirty acres of Martin's ranch is piled with
drift wood from 10 to 15 feet high. The road from
Lulle's Ferry to Galice Creek is covered with a
continued immense pile of drift (Sentinel December 31,
1861) .

Rogue River in Flood

The farms gradually recovered after the flood and
improvement of agricultural properties continued. As the
population grew, demand for a political structure increased.
Josephine County was created by act of the territorial
legislature on January 22, 1856, and the county seat was
established at Kerbyville. The county's population remained small
through much of the century, with a total number of 1,623 counted
in the 1860 federal census. By 1870, largely due to a decline in

82

mining activity, the population had decreased to 1,204 {Atwood
1984: 6) .

Much of the county's agricultural development was centered
on the broad open meadows near the confluence of the Rogue and
Applegate Rivers. Early residents of the lower Applegate Valley
received their mail at the Slate Creek Post Office which opened
September 30, 1858. (The name of the office was changed to
Wilderville on August 12, 1878) (McArthur 1992:907). The Slate
Creek School District was formed in 1859 and functioned for many
decades. During the 1880 's a new school opened at Finley Bend
along the Rogue. Named for settler Nicholas Thoss, the school was
operated on a subscription basis, which required families to pay
a fee in order to enroll their children in classes.

Most roads in the area along the river were very poor due
to rugged terrain and harsh weather. Rough intersecting trails
led from various mines and farms to supply communities. Fords and
ferries were used to transport people and goods across major
streams. The General Land Office maps of 1854-1855 indicate the
existence of trails on both sides of the Rogue River -- trails
which lead to the mines at Galice Creek.

On the north side of Rogue the map traces the "River bottom
trail from Vannoys to Gallease (sic) Creek Mines," a route which
also provided access to the farms west of Vannoy's Ferry. On the
south side of the river, the trail is described as the "Mountain
trail to Gallease Creek Mines" and follows the south and west
sides of the river, turning north at Finley Bend. Travelers on
this side of the river could cross the stream at several
locations .

At times impassable, the roads did little to alleviate the
isolation of the Rogue corridor from outside contacts. Travelers
on the Oregon-California Trail crossed the Rogue River at one of
three ferries which were situated about eight miles apart. The
closest to the mouth of the Applegate was Long's ferry, (formerly
Vannoy's Ferry) established at the place used decades earlier by
fur-traders and explorers on the Oregon-California Trail.
Following a trip to the Rogue River Valley in July, 1852, a
Willamette Valley resident described the river crossing
(Oregonian July 3, 1852) :

A mile or two before reaching the river, the road
forks; and the left hand leads to the upper Ferry,
eight miles above. We took the right and crossed at a
rope ferry, in a good boat . . . the Lower Ferry on Rogue
River is eight miles below the Middle Ferry. . .
Applegate Creek flows into the river a mile below the
lower, or Long's Ferry. On this creek are eight to one
hundred miners, as estimated...

As mining and settlement increased along the Rogue, more
dependable transportation routes were required. Upgrading trails
to roads took years. On February 7, 1860 donation land claimant
George Miller and two other settlers petitioned the County Court
for "a county road running down Rogue River from the mouth of
Gallice Creek on the south side of Rogue River to Vannoy's
ranch," (Josephine County Road Records 1:117) On May 9, 1860
local citizens petitioned the county "for a county road running
from Gallice Creek to the County line between Jackson and
Josephine County on the north side of Rogue River." The petition
was approved and three citizens named as viewers. For the next
four decades the roads were slowly improved one small section at
a time. Many years passed before a passable road would reach from
Gal ice to Almeda.

By 1875, farms were well established at Taylor Creek, at the
mouth of Jump- off Joe Creek, at Shan Creek, Limpy Creek, and
Dutcher Creek. The William Crow family, for example, settled two
sites along the river. Living first near Galice until the mid-
1870' s, the Crow family moved to the mouth of Jumpoff Joe Creek
(Keyte in Hill 1976:63). At Stratton Creek, where Indians had
butchered game thousands of years earlier, and where later
generations had fished, cattle and hogs now grazed.

Frame houses and hewn- frame barns gradually replaced older
log houses and sheds. In 1872 Hezakiah Robertson purchased a
portion of the George Miller Donation Land Claim from Edward
Neely. In 1890 Robertson's first modest dwelling was replaced by
a sturdy frame house. In 1877 Thomas Hussey settled
approximately one-quarter mile south of the river near Grade
Creek and built his first home, replacing it in 1900 with a new
frame house. Nicholas Thoss also settled a homestead in 1877 on
the south bank of the Rogue River, approximately one mile east of
Finley Bend, and soon began constructing a sturdy farm residence
(Atwood 1984) .

After 1883 and the railroad's arrival at Grants Pass,
perroanent settlement increased on the broad arable terraces
downstream from the mouth of the Applegate. Although grain crops,
fruit, cattle, sheep, and hogs comprised most production on area
farms, hops were established as a major agricultural crop on
valley land west of Grants Pass by the 1880' s. The Robertson hop
yards were located in the vicinity of Robertson Bridge and the
Finley yard lay across the river at Finley Bend. Additional hop
yards were located near the confluence of the Applegate and Rogue
Rivers (Hill 1976:7) .

In mid- January, 1890, agricultural lands again flooded when a
large accumulation of snow in the mountains and heavy warm rains
combined to flood the Rogue River and tributary streams. This
freshet wiped out almost all of the barns and houses along the
Rogue and Applegate River plains on both sides of the stream. The

g4

Jacksonville Democratic Times reported on February 6, 189 0, that
Josephine County had passed through the "greatest rain and flood
on record," with much damage done to roads, bridges, mines and
farms .

By 1900 the lands along the Rogue River had realized sweeping
impacts from mining, settlement and flood. In fifty years, most
traces of the ancient Indian culture were gone; houses, barns,
cleared fields, fences and roads now characterized the gently
rolling land along the Rogue from the Applegate River to Pickett
Creek. Roads flanked the river on both sides and ferries crossed
the stream at several places. From Hellgate to Mule Creek miners
had moved tons of rock, gravel, dug ditches, carved trails, re-
routed streams and cleared riparian areas of vegetation. Within
the Rogue Canyon small areas of settlement existed --on meadows
above Horseshoe Bend and near the mouth of Mule Creek.

Agriculture continued as a primary component of Josephine
County's economic development following the turn of the century.
Farms, however, increased in number while their overall acreage
declined. In 1900 there were 557 farms in the county with an
average size of 172.4 acres. By 1930 the number of farms in the
county had jumped dramatically to 1164 -- with an average size of
90.5 acres (U. S .Agriculture Census 1930). These changes were
reflected in the division of agricultural properties along the
Rogue River between the Applegate River and Pickett Creek. Older
residents gradually distributed property among their children or
sold out to newcomers. By 1932 for example, the former George B.
Miller Donation Land Claim had been divided between four members
of the Robertson family. Descendants of the Thoss, Vannoy, Lind
and Hussey families continued to work portions of their older
farms .

85

I" -I

Metskers Map, 1932, T 35S, R 7W, W.M.

86

At the time of the fruit and hop boom early in the century,
many long-time residents sold their land at good prices and left
the area. As economic times tightened and the Depression loomed,
many more residents left or retained only small subsistence
farms. Shan Creek, Pickett Creek, Dixie Creek and Pleasant Grove
schools enrolled students regularly until the 1930' s. With the
Depression and World War II, these schools consolidated with
larger districts and post offices closed. In 1931 the former
Grants Pass Bridge was moved approximately twelve miles
downstream and renamed the Robertson Bridge. The bridge and
improved roads meant ready access to Grants Pass and jobs in
town. Ferries, which had long operated on the river, saw rapidly
decreasing use.

During the first half of the twentieth century, floods
intermittently damaged properties along the river. In 1903, for
example, many days of heavy rain with rushing water ravaged
fields :

The farms below Grants Pass suffered serious injury,
nearly all the fences on the bottom lands were carried
away and the damage to crops, orchards and hop yards is
very great... Only meager reports have come in from the
farther portions of the county... The Grave Creek wagon
bridge was carried away. . . (Rogue River Courier January
29, 1903) .

In February, 1927 heavy rains fell over the Pacific Northwest
and flooding isolated the area for several days. Property damage
was severe throughout the area as bridges were crushed and homes
destroyed along the Rogue near Grants Pass. Among the casualties
was the Hellgate Bridge. A photograph taken at Almeda during the
high water indicates the flood's severity. The local newspaper
itemized damages:

Rogue River was swept clear of every bridge between
Grants Pass and the Pacific Ocean and five miles of the
river trail has been destroyed, according to a survey
made by the office of the Siskiyou national forest.
Flood damage in excess of $75,000 to bridges, roads,
trails and telephone lines in the Siskiyou forest was
caused by the recent high water, incomplete reports
received by the forest service from district rangers
indicate. (Grants Pass Courier March 4, 1927)

87

Rogue River in Flood at Almeda Mine, 1927

Tourism

Most of the time, however, the Rogue was not in flood and
growing interest in recreation brought new visitors to the river.
By the early 2 0th century a rapidly improving road system and
increased auto travel combined to make the canyon more accessible
to tourists. As early as 1910, a promotional brochure (Grants
Pass Commercial Club 1911) for Josephine County noted:

The Rogue River is one of the famous fishing streams of
the West. The gamey trout is here, and affords unending
delight as well as vigorous exercise, for the man who
fishes for its enjoyment. . . .

Soon the beauty and wildness of the Rogue Canyon was broadly
recognized and residents and visitors made use of the trails to
explore the back country. The Pacific Highway, routed through
Ashland, Medford and Grants Pass, brought thousands of visitors
through the Rogue Valley. Tourists could reach Galice by leaving
the Pacific Highway near Merlin, Oregon and explore previously
unreachable areas of the country.

88

During the earliest years of the century recreational
activity was intertwined with work and food acquisition. By the
1920' s, however, most of the visitors who came to the Rogue were
recreational anglers and outdoor adventurers bound for a trip
downriver. The river's excellent fall fishing season had long
been advertised in an attempt to encourage settlement in the
Grants Pass area.

Zane Grey, writer-adventurer, who publicized the river's
wild beauty, brought wide recognition to the river with the
publication in 1924 of Rogue River Feud, and in 1928 with his
personal account. Tales of Fresh Water Fishing. On a 1925 fishing
trip down the Rogue, Grey recognized the beauty of Winkle Bar,
and the following year bought it from a prospector who held it as
a mining claim. In Tales of Freshwater Fishing Grey described
his new acquisition:

The rushing river at this point makes a deep bend round
a long oval bar, with rocky banks and high level
benches above, and both wooded and open land. Here it
flows through a lonely valley set down amid the lofty
green mountain slopes. A government forest trail winds
out some twenty miles to the nearest settlement . Far
indeed it is across the dark Oregon peaks to railroad
or automobile road! (Grey 1928:257).

Grey's enormous popularity with the public and his frequent
writings about the Rogue focused world-wide attention on the
canyon. Between 1910 and 1924, a Zane Grey book was included
every year but one on the bestseller list. Many summer visitors
to the Rogue Canyon came because they had read of it in one of
Grey's books (Korbulic 1992:5)

During the bleak mid-Depression years, a Grants Pass
newspaper enthusiastically publicized the river canyon (Grants
Pass Daily Courier, April 3, 1935) :

The Rogue River claims greatest attention as it has
brought to itself a world-wide fame and glory, and is
acclaimed as the premier fishing stream of the
continent... It is a scenic marvel, a joy alike to
amateur photographer, to canoeist, the motor boat
enthusiast, the lover of water sport. But the greatest
lure of the Rogue is for the angler. Fisherman come
from all parts of the United States and from foreign
countries to match skill with the leaping steelhead or
to do battle with the lordly Chinook.

During the 1920 's and 1930 's fishing lodges appeared in the
Rogue Canyon and adjacent areas. Some of the best quarters could
be found at "Speed's Place on the Rogue." This attractive, rustic
lodge was developed in 1928 near Galice at the former Galice

89

Consolidated Mine headquarters. The inn's main building had
served years earlier as cookhouse, dwelling and office for the
mining company.

In 1928 Wallace Robertson, known informally as "Speed, "
purchased the Gold Leaf and Genevieve placers which included the
old mine company buildings. In 1929 Robertson constructed seven
cabins near the main house and opened "Speed's Place on the
Rogue." The two-story board-and batten dwelling now served as the
main lodge while the cabins provided sleeping accommodations for
guests .

* ti :3#-f# *'

"Speed's Place" on the Rogue, 1987

"Speed" Robertson also served as a popular river guide. As
an expert fisherman he accompanied Zane Grey on several
downstream expeditions. In February, 1933, Robertson sold the
resort to Fred MacFarlane who transferred title to his daughter
and son-in-law, Zora and Pat Gallagher. Reasonable rates, good
food, and comfortable cabins brought many visitors. In January,

90

1937 a reporter for the Grants Pass Bulletin, described her trip
to Galice and recommended a stop at "Speed's Place:"

Within a few miles we come to Frank
Carpenter's store... Next is "Speed's Place,"
a fine attractive auto camp. Probably you
will want to stay in one of their cabins,
providing you want to stay and see the
country (Grants Pass Bulletin, January 29,
1937) .

The Gallaghers operated the resort until the late 1930 's
when the Depression required them to seek work elsewhere. The
main building and three guest cabins, surrounded by well-kept
grounds, remain visible along the Merlin-Galice Road. Speed's
Place on the Rogue, provides a tangible link to early twentieth
century tourism, a significant part of the Rogue's story.

CCC Camp at Rand, Aerial View, ca. 1935

91

During the early 1930' s the Depression brought difficult
times to the Rogue River Canyon and residents mined, cut wood and
grew their own food to survive. In 1933 the Civilian Conservation
Corps (CCC) opened a camp near the forest guard station at Rand
in 1933. The Corps, organized to provide work for citizens
needing economic assistance, the Civilian Conservation Corps
crews contributed substantially to physical improvements within
the Canyon area. Under supervision of the U.S. Forest Service,
young men from this base constructed miles of truck trails,
fought fire, and built bridges within the remote reaches of the
river canyon. Soon after the Civilian Conservation Corp opened
its Rand camp, the Grants Pass Courier (April 3, 1935) reviewed
the camp's history and accomplishments:

Camp Rand is on the meadow where pioneers settled, on
the place where the town of Almeda later supplied the
miners and trappers. Camp Rand now houses 200 men from
the rest of country who are engaged in building roads
and bridges .. .September 30, 1933 was the date of first
breaking of ground for Camp Rand, as eighteen men
started the camp...

Bridge Over Rogue River at Grave Creek

CCC crews spent much of their time on road construction and
introduced important truck roads into the isolated area.
Substantial portions of road were completed on the Almeda Road

92

No. 340 and on a route from Saw Mill Gap by Cold Springs to the
West Fork of Cow Creek railroad crossing. CCC crews also worked
on the road from Ninemile Spring to Mule Creek, completing four
miles in 1934. With WPA crews' assistance, the road reached Bald
Ridge and the following year was completed to the Rogue River
Ranch at Mule Creek. (Grants Pass Courier. September 1, 1934) .

Between late 1934 and 1935 CCC crews completed construction
of the 344 foot long suspension bridge over the river near the
mouth of Grave Creek. The local newspaper described the benefit
to tourists, stating, "Completion of the bridge will open a new
scenic loop for sight-seers. They can go by way of Merlin to
Galice and on down to the bridge and return to the Pacific
Highway via the Grave Creek Road. " (Courier 1935) George Morey,
member of the CCC unit stationed at Rand, recalled the massive
bridge construction project:

During the winter of 1934 and 1935, at the base camp,
the road was passable as far as the mouth of Graves
Creek. A new swinging bridge was under construction
across Rogue River at this point. This bridge was
designed for heavy traffic of that time. By the late
spring of 1935, the bridge was far enough along to
allow traffic to cross the river. The old trail
bridge, used for many years by the pack stock, was
dismantled and packed the one and one-half miles out to
the road (Morey n.d.)

Most CCC men left the area by 1941 to serve in World War II.
Miners either joined the military or sought jobs in war-related
industries. Along the Rogue as in the rest of the country,
development slowed as the war effort claimed residents'
attention. The post-war years would bring sweeping changes to the
river .

— ■ mmmmmamummmmmmmm.

r>MiM..w'«wiQii»wiui )(.i:iiiiiiiiinii«iii,ii II, iiiu mijiiiillluiiIlliiWLlluillMiiUliMilMMillilil^llli""*^^

^:s^.V5^:i ,A^#u^ ?:0isZSTj

''XHHH^ <^ 'KtWH^ -^Ufi

Grave Creek Bridge Plaque, 1935

93

Figure 8. Map of the Rogue River from Grave Creek to Horseshoe
Bend

94

VII. GRAVE CREEK TO MULE CREEK

Thousands of years ago a band of people gathered in the late
autumn sun on a gently sloping terrace above the river. At this
site, near what is now called Mule Creek, some of the inhabitants
fished while their kinsmen butchered freshly-killed game taken on
the upland meadows. After several harvest seasons here, the band
would soon move on to new hunting and foraging grounds. In the
autumns of later millennia a different people assembled here to
build shelters, cook game and grind acorns. Then after many
generations, the Indians were gone from the canyon. In 1855 they
camped their last autumn at this place by the river where already
the ground was trampled by miners and the stone tools scattered
over the ground.

Mining

With the Indians gone, miners freely roamed the Rogue canyon
and soon men and equipment were scattered along Grave Creek,
Whisky Creek, China Gulch and Mule Creek. Like the fur- seekers
before them, the miners intently extracted the river' resources.
John Mickey, who worked in the Rogue Canyon from 1856 to 1859,
was typical of the men who sought wealth from the ground. He and
his partners packed into the canyon from Galice. They staked
their first claims on Whisky Creek. Here Mickey (Mickey in Atwood
1978: 185) wrote:

We spent a little time on Whiskey Creek prospecting,
found a little trace but not much. We were camped at
its mouth on the banks of the Rogue River. One day we
built a raft and crossed over to the south shore,
prospected and found some gold on a bar just opposite
Whiskey Creek. It looked promising, so' we moved camp,
set up a whipsaw and started building sluices... We all

went full partners Within three weeks we were

sluicing, having started sawing on August 12, 1856.

Some miners straggled into the canyon from the Rogue Valley,
others followed game or Indian trails from the Grave Creek or Cow
Creek drainages. Miners struggling to reach Mule Creek from the
west in 1856 had a grueling trip:

The station [Big Bend] is situated on the north side of
the river about 20 miles below [Big Meadows] . . . There
is a trail down to the station on the same side,
keeping the bluffs and crossing the mouth of what is
called the John Mule creek, which is within fourteen
miles of said station. The trail is not a very good one
and mules should not be packed with more than 150
pounds to go over it .... (Crescent City Herald. March 5,
1856) .

95

At Whisky Creek John Mickey and his partners prepared for winter,
While some of the company built shelters, Mickey headed out for
supplies. His diary indicates that storeowners were impressed
with his financial resources:

. . . The boys were building houses now, expecting to
stop here for the winter. They sent me out for

winter's provisions $1400 worth in all. The

merchants thought I was a storekeeper. The boys had
made all this money in about two months, besides
building three log houses. (Atwood 1978:185)

Cabin Near Rogue River, ca. 1930

96

In spite of weeks of hard work at Whisky Creek, the company-
found better prospects a short distance away and concentrated
mining efforts there. John Mickey's diary reveals the back-
breaking work required to meet their goals:

We had been prospecting a bar down the river a mile
below our present one . We thought it would pay to put
water on. Its name was Tyee Bar. The upper bar was not
large enough for all of us, so on the 24th we began
building a sluice a mile long, to Tyee Bar. . .We were
about ten months building the flume to mine Tyee
Bar. . .The flume crossed canyons, and stuck up on poles
twenty- five feet high, and part of the way it was on
the mountainside, so we had to lash ourselves by ropes
to the trees to keep from falling down the
mountainside. . .Tyee Bar is half a mile long and 300
feet across, from five to 30 feet deep, mostly round
boulderd from a hen's egg up to a ton. . . .We had all
been living in the three houses on Upper Bar. Now we
sold them and built a new house on Tyee Bar. . . (Atwood
1978:187)

Only a few single miners, men like John McGraw and John
Adams, lived permanently in the canyon during the 1860's and
early 1870' s. The few small terraces above the river were just
wide enough to hold a shelter and equipment and most of the men
moved frequently. Tom East was typical of the solitary men who
worked at Mule Creek in the mid-nineteenth century. At various
times he worked near Galice, and later had claims on Grave Creek,
Alder Creek and at Brushy Bar. When he died in July, 1897 after
fifty years in the river canyon he had four creeks named after
him.

Few men filed claims -- most panned out what they could by
sniping along the bars and exploring the creek banks. One of the
earliest recorded claims was taken in 1865 opposite the mouth of
Mule Creek, "on the River, in accordance with the John Mule Creek
Mining Laws" (Mining Claims A:9). By 1870, the federal census
counted about fifteen prospectors in the Rogue canyon between
Grave Creek and Mule Creek. These hardy men fished and hunted
game as people before them had done, but they also purchased food
at Ellensburg (Gold Beach) , Roseburg or Galice, hiking out each
season to replenish dwindling supplies.

Chinese miners worked many locations in the Rogue Canyon
after 1855. Names such as China Bar, China Riffle and China
Gulch are reminders of their presence. Chinese miners continued
to work placer deposits on Rogue River terraces long after Euro-
Americans had extracted the more easily mined gold. In 1870 the
federal census taker found a small number of Chinese working the
bars near Mule Creek.

97

When the census was taken again in 1880 there were almost no
Chinese miners left along the river. Following the passage by
U.S. Congress of the Chinese Exclusion Act in 1882, in which
Chinese were declared ineligible for citizenship, almost all
Chinese disappeared from southern Oregon. In recent years Chinese
objects have been discovered at Tyee Bar, Almeda and Little Windy
Creek. These articles are important evidence of the Chinese
miners who worked the river bars (Gray 1995:6) .

Chow Long lived for some years just upriver from the mouth
of Mule Creek. Ivin Billings, (Atwood Illahe:98) long-time
resident of the Mule Creek area, recalled the cabin where the man
for whom China Riffle and China Bar were named lived out his
life.

Sketch of Chinese Miner

98

You have to go up the trail about a mile to get [there]
There's no sign of the cabin now .. . [or] where he was
buried. The floods have washed it off and the bank is
caved in. He was buried right in front of the cabin;
right on the bar.

For several generations mining in the Rogue Canyon followed
a similar pattern as miners continually moved rocJc and gravel,
using piclcs, rockers and sluice boxes to extract the gold.
Although the richest gold deposits were exploited between 1852
and 1860, mining continued although productivity gradually
diminished through the years.

After the placers had been skimmed by hand, hydraulic mining
began. Crews dug endless miles of ditches to the sites.
Hydraulic mining diverted water from mountain streams into
ditches and carried it across ravines and through flumes before
dropping it into a pipe and nozzle with which to blast away
gravel and soil on bars and slopes. The concentrated water stream
washed entire hillsides down and the loose soil was run through
ground sluices to extract the gold (Atwood 1978) .

John Mickey's diary marveled at the water's power and makes
it evident that hydraulic principles were in use on the Rogue
River as early as the 1850' s.

... We intend to ground sluice till April, then clean
up till about July 4th. The Hydraulic is a great
invention. It does the work of twenty men and is not
half as dangerous as pick and shovel. You just stand
back and play the water under the bank till she caves
in, then wash it off, then repeat, and down she comes.
... It is wonderful what power there is in water
confined. . . . (Atwood 1978:190) .

As surface deposits were exhausted, new gold deposits were
found buried far beneath the large gravel bars -- bars which
could be a half-mile in length and contain "upward of forty feet
of gravel atop the rich paydirt (Beckham 1978:106) ." Rapidly
advancing technology and improved hydraulic methods made it
possible to retrieve the wealth from these deep levels. By 1875
hydraulic methods enabled miners to extract much of the low-grade
deposits remaining in the Canyon.

In 1875 Asher and Solomon Marks, Roseburg men, mined several
locations along the river in the vicinity of Mule Creek. Packing
in extensive equipment and hiring sizeable work crews, the
company constructed a high trestle bridge and flume across Mule
Creek as well as a long ditch to their mine on the river. (Atwood
1978:100) .

99

Three years later, in 1878, when Army Corps of Engineers
Lieutenant Philip Eastwick traveled from Grave Creek to Big Bend
while investigating potential navigability of the Rogue, he
undoubtedly observed Marks and Solomon among the many miners
working sites along the river. Eastwick reported (Atwood
1978:231) :

.... Along the banks of the river I found many miners'
cabins, and evidence on the bars and rock banks of the river
of recent mining operations, and in a few cases of quite
extensive hydraulic mines... When water is abundant for
mining purposes, the gold mining carried on at many places
on this part of the river and its tributaries is very
extensive... Many small companies of miners... mine on a
small scale along this entire division.

Up-to-date hydraulic methods required more money than had
older simpler techniques, and new enterprises were often
supported by outside investors. The isolated mine locations
demanded freighting and packing large, heavy equipment long
distances. Many men entered the river canyon to pack supplies
and to labor in the mine work camps . The arrival of the Oregon
and California Railroad at West Fork on Cow Creek in 1883 brought
a supply center closer to the Mule Creek area and substantially
reduced the distance over which the massive amounts of equipment
and supplies were packed.

An 1881 newspaper report in the Port Orford Post indicates
damages from high water but also describes the extent of mining
activity in the canyon that was interrupted only briefly when
floodwaters sent debris splintering over the rocks:

At the foot of Battle Bar, Russell and Co. are working
with good prospects of success. Low bars in general
along the river have been sluiced out; boxes, flumes,
ditches reservoirs, cabins etc. went down to the sea
during the freshet.

Miners remained busy in the canyon through the final two
decades of the nineteenth century. In 1887 the Grants Pass
newspaper observed (Courier, May 13, 1887), " There is extensive
work going on in the mines on Whiskey Creek, Josephine County.
Forty or fifty claims are being worked and good returns are
realized." Three years later miners were discouraged when
another major flood caused "great damage to the mining interests
by the washing out of reservoirs and dams, breaking and filling
up of ditches, losses of flumes... etc. The work of months of
preparation has been undone in many instances" (Democratic Times
February 6, 1890)

Tyee Bar was extensively mined during the later years of the
nineteenth century. J.S. Diller noted years of activity at Tyee

100

Bar Mine -- the place where John Mickey and his company had met
success. Diller (1914) wrote:

"The Tyee Bar placer on the left bank of Rogue River,
about one and one-half miles below Whisky Run, although
not large, embraces a number of acres. Much of the bar
was worked over years ago and reported rich."

By 1880 placers at the mouth of Whisky Creek enticed large
numbers of miners to the district. Among the placer miners who
had settled in for a season were men who wanted a sturdy,
permanent dwelling. They selected a site for a cabin on Whisky
Creek about 1000 feet above the river and began construction of
twenty by twenty foot residence -- larger than most in the
canyon. After days of cutting and notching logs, the men laid
the logs for the walls and split piles of sugar pine shakes for
the roof .

V/hisky Creek Cabin, Notched Logs

About 1890 a flume ditch was begun one-half mile up Whisky
Creek from the cabin and completed to a point near the structure.
In 1905 the flume were substantially extended to a location about
sixty feet above Whisky Creek Bar. For years the flume ditch
provided drinking water at the cabin and water power for
hydraulic mining on the river.

101

Whisky Creek Cabin, ca. 1925

P.H. Rushmore filed the first recorded mining claim on the
Whisky Creek cabin site on January 5, 1917. Hanson A. Whiteneck
bought the claim in 1918, enlarged the cabin and added a full
length porch on each side. He installed a fir floor and divided
the interior into two rooms -- one to serve as the living
quarters and the other as a storeroom. In later years new
structures were added to the property, including a woodshed,
fruit room, shower and auxiliary sheds.

In 1948 the claim was sold to Mr. and Mrs. L.M. Nichols who
retained ownership until the property was acquired by the United
State government . The Whisky Creek Cabin was entered into the
National Register of Historic Places in 1975 (Abbett 1973) .
Flood, fire, demolition and neglect have eliminated most old
miners' cabins from the Rogue Canyon and the Whisky Creek Cabin
remains one of the regions most distinct, rare resources.

For much of the late nineteenth and early twentieth century,
hydraulic placer mines operated on several of the bars and
benches above the river level. At Tyee Bar, Black Bar, Little
Windy Bar and Horseshoe Bar, crews steadily blasted the soil,
churning tons of rock and gravel. The Horseshoe Bar Mine, for
example, encompassed two twenty-acre claims within ten feet of
the river level. Water was supplied to it by a mile of ditch and
a pipe which was bridged over the river at an approximate
elevation of ten feet. Farther downriver hydraulic mines
operated on Winkle Bar and Battle Bar (Diller 1914) . In 1891 John

102

Billings operated a sizeable placer mine at Red River Bar a half
mile downstream from Marial .

In June, 1906 the Red River Gold Mining and Milling Company,
an Indiana based corporation, purchased several claims at the
mouth of Mule Creek. Much of the land they bought had been
previously mined by the John Billings and others. The Red River
Gold Mining Company operated for six years before closing in
1912. During this brief period the company took between eight and
ten thousand dollars worth of gold out of the soil ( Mineral
Resources 1916) . Private interests ran the mining operation for
about ten additional years. The later organization constructed an
extensive ditch and flume system which brought water three and
one-half miles in a four- foot flume from Mule Creek and built
several trestles (Beckham 1978:108).

Ben Corwin, whose father William Corwin, was a part owner in
the Red River Gold Mining Company after 19 06, recalled the long
flume and high trestles:

They'd cut [lumber] in the day time, stack up their wood,
and in the evening they would turn the water into the flume
and float the lumber down to where they were building. They
had men down there picking the lumber out and setting it
aside. Next day they would go ahead and build another
section. . . The closest place to the ground was three feet
and the highest place was 310 f eet . . . (Atwood 1978:104).

Red River Gold Mining Company Trestle, ca. 1912

103

When the venture was finally abandoned years later, portions
of the flume and trestles were torn down for lumber or allowed to
disintegrate in place. Pieces of the trestle can still be seen on
the hillside above Mule Creek.

By 1920, with low returns and high operative costs, large
hydraulic mining operations gradually decreased along the river.
Although the Depression years would bring renewed activity as the
price of gold increased, much of that effort would be devoted to
individual placer or lode mining efforts. The impact of early
twentieth century hydraulic operations is still evident in the
Rogue corridor in huge piles of mine tailings, ditch remainders,
and scattered lumber left from improvements and wooden flumes.

Hardrock or lode mining, which had begun in both the Galice
district and the Mule Creek district before 1900, developed
rapidly after 1900. Lode mining, like hydraulic mining, required
substantial capital and frequently demanded the backing of
corporate groups. Processing ore from lode mines required that it
be drilled or blasted from the vein, loaded into cars and taken
to the surface.

In early years the gold-bearing ore was extracted by the use
of an arrastra, a grinding method employed in the Americas since
early Spanish exploitation of mines in Mexico. The arrastra
pulverized ore between horizontal millstones which were turned by
hand or by oxen. More modern technologies eventually reached the
area and by 1900 more sophisticated mechanical and chemical means
were available to extract and process gold (Beckham 1978:110).
Stamp mills employed heavy iron pestles (stamps) which crushed
the gold-bearing ore prior to chemical processing. Gold recovery
was accomplished by any of several processes, depending on the
ore's mineralogical character. Amalgamation, flotation and
cyanidation comprised major processing methods (Brooks and Ramp
1968:33) .

104

George Billings at Mule Mountain, 1930

In the Mule Creek area lode mining was primarily conducted
in the river corridor at the Mammoth Mine, the Marigold or (Tina
H) Mine, the Mule Mountain Group and the Paradise Mine, which
were all established between 1895 and 1905. Although Brooks and
Ramp assessed lode mine production in the Mule Creek District as
relatively modest, the years of effort at those mines was
intense. They state that, "[Production] has come mainly from
early-day gold-placer operations on lower Mule Creek, Rogue River
bench gravels below Mule Creek and along the upper West Fork of
Cow Creek" (Brooks and Ramp 1968:189-191).

Gold-bearing quartz mines were distributed in several areas
near Galice and farther downriver, occurring in veins of varying
widths and classes of rock. In summarizing mineral production in
the Galice area between 1854 and the early twentieth century,
Stephen Dow Beckham (1978:115) writes:

Mining commenced in this area about 1854 but tapered
off within a decade after the richest placers had been
worked. . .Quartz mining. .. included fairly extensive
milling and processing activities at the Gold Bug,
Golden Wedge, Sugar Pine, Almeda, Oriele, and Bunker

105

Hill Mines. Between 1854 and 1912 an estimated $3
million was extracted in the area.

Although placer and lode mining continued through the 1920' s
the Depression brought an influx of people into the canyon. As
other sources of income declined these newcomers were enticed by
an increase in the gold price in 1934 and the chance to carve out
a subsistence living in the Rogue Canyon. Shelters could be made
of poles, split shakes and discarded materials and game, fish,
and plant foods would keep people alive.

Unidentified Miners, Rogue River Canyon

Many individuals and families made modest livings on the
creeks and terraces of the Rogue River. In addition to sniping
for gold, these residents occasionally found employment with the
U.S. Forest Service packing or fighting fire. Fishing and
hunting sustained many people who built simple shake and pole
cabins or stayed in abandoned shelters.

J.C. ("Red") Keller, born in Grants Pass in 1909, hiked into
the river canyon 1934. He lived over fifty years in the place he
called home. Skilled at cabin location and construction. Red
Keller used materials he found or could fashion from the trees
and brush (Atwood 1978:175):

First thing I look for in a cabin site is whether I can
get water close. Then pick the site high enough. I have
to pick it where I know I can get wood. . . What you need
is a good big dead snag you can fall. I like my site
where I can get the sun. That's what I like about
Horseshoe Bend. You don't get the sun real early, but

106

you get all the late sun. Winter, if there was any sun,
you got it .

Most people down in there just built their own cabins.
One person could build the pole and shake cabin... You
stand your poles up and then you put your poles
crossways to put your shakes on, and your roof too...
Nothing real heavy, but it'll stand up for a long
time... Most cabins had dirt floors because no air will
come in when its dirt... Sometimes you cooked outside.
One place they just took tin cans and sleeved them
together and that was the chimney... You'd be surprised
how you can fix up a little cabin and have it looking
pretty neat in there.

Red Keller, 1976

Red Keller hunted and fished, purchased eggs and vegetables
when he could find them and never went hungry. He explained,

I knew how to shoot . . . I knew how to fish. Generally I'd
have fish twice a week. You could hunt deer in the
winter and can it. You bone it, cut it and pack it in
the jars. Put your lids on and let it come to a boil.
Just so its boilin' for three hours . . . I usually took
eggs in. There 'd be venison steak and potatoes. I'd

107

make biscuits. You don't need any grease for them.
Watercress took the place of lettuce. I never tried to
grow anything because I never figured on being in a
place long. Back up in the hills there was a few
permanent homesteads. You could buy eggs and stuff out
of the garden.

Miners drifted season to season through the hills and along
the river's edge. Their main outside contacts were the packers
who brought in supplies. For the most part they left no written
trace of their existence. A few gave their name to a creek, bar,
or gulch, more lie in unmarked graves near cabin sites or along
the trail where they were found.

Price Copsey at Cabin, ca. 1915

Red Keller explored many mining locations through the years
and lived in many different cabins. As he explained:

Thirty-four was when I first moved in there. I walked
down from Almeda . That was the end of the road. I took
the trail down about a mile or so. The first winter I
stayed at Russian Creek and then I went down to Black
Bar to the old Copsey Cabin. I stayed at Galice one

108

whole winter. I stayed at Slate Slide below Howard
Creek one winter... Mostly I just camped on and off.

I've worked up and down the river, from Kelsey Creek to
Slate Slide or Bronco Creek. I worked back and forth
and kept on hunting crevices. I always had an ax and a
shovel, and usually some picks with handles, but small
picks ...because we'd move from place to place. After
you'd spotted all the places you figure would pay, then
you'd go up and get your long toms and screen and start
moving things closer ... Sometimes when we left we'd just
leave the old equipment. We'd just build a new sluice
box. . .It was easier to build new than trying to cart it
around .

Lou Martin, 1976

109

Lou Martin, who came to the Rogue River in the late 1920' s,
mined first at Howard Creek and later on Rum Creek, Tyee Bar and
at Rainey Falls. He relished the isolation:

I found out I could be independent, as long as I stayed
in a mineralized area, in placer country. I could make
my own livelihood. Don't figure you're going to get
rich, 'cause nine times out of ten, you're not... You're
independent, you punch no clock. If you don't like the
way things is going, you pick up your bed and you move
across the hill, or to another creek. You have to do a
certain amount of work to eat up here. . . Then if you
want to play, you play (Atwood 1978) .

There were lots of signs of old miners when I went in
there. Especially where they had their camp. Hundreds
of them. No matter how much a miner prospects a country
there's always a chance of another man coming in and
making it. . . On Howard Creek I could still see where
they'd piled their rocks up. There was a lot of old
ruins on Howard Creek when I first came in. . .

Sometimes it would be three or four months that I
wouldn't see another human being. As long as you're
working, it don't make any dif f erence . . .when you mine
for yourself you don't put eight hours in. It's twelve,
fourteen hours a day. . .

Miners at Mule Creek, ca. 1935

110

Bob Simmons, Jack Bowen, Bill Brockman, and Clayton and
Clifton Sanderson were among others who lived for years in the
canyon. The latter were miners and good engineers. In 1939 the
twin brothers poured the foundation for their home on the river
near Galice, built a shop and developed their own power system.
L.A. Damon, Charley Lewis, Arnold Ezard, Charley Tucker, Shorty
Borden and Henry Barlow were among the men who built pole and
shake cabins in the Mule Creek Canyon during the Depression
years. A few of them are buried in the small graveyard at Marial

111

Figure 9. Map of the Rogue River from Horseshoe Bend to Mule
Creek

112

World War II brought an end to most of the subsistence
living created by the depression era miners. The sale of gold was
halted in 1942 and military service or job opportunities in war-
related industries took many of the younger men away. Most never
returned to the Rogue to live. In the years following the war new
roads eased the long isolation of the area. Tourism increased as
anglers, hikers and boaters flocked to the river canyon. The few
individuals who wished to continue their life "in the hills"
managed to do so with ingenuity.

Settlement

People who settled on the few places in the Rogue River
Canyon which would support a house, crops and grazing stock, were
isolated from the rest of the world, but not from each other.
They worked long, hard, bone-wearying hours and most of their
energy was spent acquiring food, shelter, and clothing. These
people survived with very little money in a "subsistence"
economy, in which they grew, hunted, fashioned or bartered for
the bare necessities of life.

Walter and May Criteser

Below Grave Creek, the rugged Rogue River Canyon was
sparsely populated. Few settlers are known to have farmed the
occasional sloping meadows or small flats above the river. In
1878 Lieutenant Eastwick noted the primitive nature of the
country (At wood 19 78) .

Back from the river and in the smaller valleys shedding
to the river, as well as up to the slopes of the
flanking hills, an occasional settlement is found...
their manner of living is generally rude and primitive,
and but few of the luxuries or refinements, or even

113

comforts of civilization have yet found their way among
them. Whatever land is found fit for cultivation it
produces abundantly of all manner of fruit and
vegetables... but little grain is raised, and this,
together with the other products of the soil, is
consumed at home. A principal occupation of the
settlers is the raising of sheep and cattle, of which
large herds are grazed...

Miner Henry Rosenbrook, known as "Dutch Henry, " was among
those living in the canyon when Philip Eastwick passed by.
Rosenbrook took a homestead at Little Meadows near the Curry
County boundary where he raised cattle in the hills. To reach
supply centers to the north, he laid out an extensive trail
system. Known as the Dutch Henry Trail, the route led out along
the ridge between Meadow Creek and Kelsey Creek to Glendale,
Oregon, following the divide between the Cow Creek and Rogue
River drainages. The trail was regularly used by others seeking
supply routes from the canyon. Rosenbrook, who had arrived in
the Rogue Canyon in the early 1860's, lived above the river until
his death early in the twentieth century.

; /' '
0 / ^

^^^u-^v

'■^\ >w "^-^ ^--^

Rogue River Between Horseshoe Bend and Winkle Bar

114

In contrast to the few meadows existing
sloping land near the mouth of Mule Creek of
settlement opportunities. Long a hospitable
area had supported people intermittently for
It was here that the Billings family establi
community which survived for decades. Isolat
terrain and unnavigable river, these people
exploited the resources around them, includi
and gold. They raised their own stock and f
their own tools, and traded or bartered for
manufacture (Winthrop 1995) .

upriver, gently
fered promising
living site, the
thousands of years .
shed a ranch and a
ed by the rugged
and their neighbors
ng fish, game, fuel
ood, crafted most of
what they could not

George, Sarah and Ivin Billings, ca. 1900

115

The George Billings family picked wild berries, raised their
own fruit and dried it for winter use. They hunted deer, elk, and
bear, depending largely on venison as meat. Hunting provided a
diversion from drudgery as well as food for the table. In huge
garden plots they raised corn, squash, tomatoes and onions. Ivin
Billings recalled (Atwood 1978) :

We grew everything... Lots of times we'd raise winter
crops too... By spring those turnips would be big. We
fed that to the stock and to ourselves and to the
miners. Carrots and beets would stay in the ground all
winter. We always had all kinds of fruit. There were
three or four peach trees set along the ditch. My
mother grafted trees in the front yard and crossed
seedling apples down here... We made bacon out of
bear. . . We'd smoke about three or four gunny sacks full
and store that in the upstairs attic. We raised hay,
oats and vetch. We sold hay and would pack it around.
We bought coffee in wooden drums. We'd grind it and
sell what we didn't use to miners and farmer around.

In 1881 E.H. Price constructed a small cabin at the mouth of
Mule Creek, raising the first known Euro-American dwelling at the
location. Although Price eventually relocated downriver at Big
Bend his cabin remained standing for several years and housed
occasional occupants.

Adeline Billings, ca. 1915

116

In 1890 George Billings worked at the hydraulic mine run by-
Solomon and Herman Marks at Douglas Bar. The Indian Allotment Act
of 1887 (the Dawes Act) , amended in 1891, reduced Indian land
holdings at Siletz Reservation, but allowed those of Indian
heritage to apply for title to homestead land (Gold Beach Gazette
February 2, 1894; Beckham 1978:83). Billings, whose mother,
Adeline, was a Karuk Indian, was eligible for an allotment. In
1895 he took advantage of the homestead opportunity and settled
permanently at Mule Creek with his wife, Sarah Ann and son, Ivin.
They moved to the meadow where Elijah Price's old cabin stood and
lived temporarily while Billings built a larger cabin.

The Billings family depended on each other for help during
illnesses or when injury occurred. Adeline Billings, who brought
her knowledge of medicinal plants with her to the Rogue Canyon,
was often called upon for help. Her granddaughter, Marial
Billings Akesson remember her curative skill:

Grandmother took care of everybody; babies,
everything... After a man was bitten by a rattlesnake,
she packed his leg in mud. She went out and got a
certain kind of mud and packed his leg. He lived. Her
mother had taught her about the use of herbs, roots,
and leaves. She knew the medicinal qualities everything
had. I remember one time I got awfully sick with bad
cramps in my stomach. Grandmother went out and brought
in what she called smart weed and made tea. The next
morning I was out cavorting like a young
squirrel . (Atwood 1978)

Baskets Made by Adeline Billings

111

Adeline Billings provided medical care for many canyon
residents. In addition to caring for her family and for others
she made baskets which she sold to miners. George Billings built
most of the tools needed for daily existence on his ranch. Ivin
Billings recalled his father's skill:

My Dad made tools right here. He could build about
anything he wanted. We made chisels and bars; bars to
drill rock; picks and re-steeled picks. We'd take the
eye after the pick would wear down till you didn't have
any steel left. We'd weld a piece of steel on to each
end of the pick and make a new pick out of it. . . My Dad
made all his packsaddles. His lash hooks were made out
of oak. They were better than boughten ones... We never
throwed anything away. Everything was saved.

In 1883 the Oregon and California Railroad construction
reached a point on the West Fork of Cow Creek, offering a closer
and more convenient supply point for residents along the Rogue
River and through the surrounding hills. George Billings realized
the possibilities for acquiring equipment and supplies at West
Fork. He and others soon cleared a trail through the mountains
which led them directly to the West Fork Station. The twenty-one
miles to West Fork from Mule Creek could be traveled in seven
hours with a loaded mule train and proved much more efficient
than the old steep and difficult trail over the Big Meadows to
Camas Valley and Roseburg,

Pack Train, Mule Creek

George Billings began a packing business and opened a small
store in his cabin. Packing regularly out to West Fork he
purchased equipment and supplies and loaded them back to his
store to sell to miners and settler families. Blacksmith tools,

118

farm tools, flour, sugar, kerosene, furniture, coffee, beans, and
fabric were packed the twenty- two miles from the railroad. Ivin
Billings recalled:

From the time my folks first started the store they
would go out to West Fork and buy supplies for the
whole country. We'd spend the latter part of the whole
summer packing. Day after day, after day, from West
Fork and go back the next day. every other day we'd
come in with a load.

We packed to the Mule Creek Mines, Blossom Bar Mine,
Douglas Bar Mine. We'd always get our supplies at West
Fork in the Fall of the year. We'd pack a ton of flour,
a ton of potatoes... They got everything in the fall
before the snow fell. If the snow got too deep, you
couldn't get out.

Railroad construction encouraged postal
canyon. After over a decade of effort, a mai
established up the Rogue River from Wedderbu
In addition to postal facilities at Illahe
1895) and Agness, (opened 16 October, 1897
at Marial on January 29, 1896. The mail was
Fork to Agness by mule team and then transpo
Rogue River. In October, 1903 a post office
with Thomas Billings as postmaster. The new
Marial for the postmaster's young daughter (

seirvice in the river
1 route was
rn to the Big Bend,
opened 7 February,
a post office opened
carried from West
rted by boat down the
opened at Mule Creek
station was named
Price 1967 :10-11) .

George Billings Homestead, (Rogue River Ranch) 1915

119

About 1900 George Billings began work on a larger home for
his family, using local materials including lumber milled from
trees cut from the homestead. The commodious dwelling was
completed about 1903. Now known as the Rogue River Ranch, it lies
on gently sloping ground along Mule Creek. The main dwelling,
with painted wood siding and wood frame windows, painted red, is
highly visible on the broad meadow. Five years after first moving
to Mule Creek, George Billings had a substantial frame dwelling
of local whip- sawed lumber. Ivin Billings described the house's
construction:

Main house, boarding house, trading post, call it what
you will. Lumber for the house was all made by hand. We
used ponderosa pine growing out on the flat, where the
three apple trees are now -- between the two pastures.
Had five saws going at one time... ten men. Whipsawed it
right here. The outside was built in a couple of years,
but the inside took six or seven years ... the glass for
the windows was packed over from West Fork by horse.
The furniture, chairs, sofas and rockers was all
handmade by Father .

In 1908 the Billings family built a large barn to house pack
mules and equipment. The upper level of the barn consisted of
one large room with plank floor and walls. Numerous social events
were held here, including dances and church services. The
latter, conducted by George Billings, eventually earned the
building it's present informal name, "the tabernacle." Miners,
packers, and homesteading families crowded into the barn for
parties and voted there on election day.

Billings Family and Friends in the "Tabernacle, " 1915

120

George and Annie Billings' family opened their doors to
travelers through the Rogue country. Charging a dollar a day for
three meals and overnight lodging, they could accumulate enough
money to buy the things they could not make themselves . The house
held the family of five as well as the guests. A large space
upstairs was divided into small cubicles separated by sheets of
heavy unbleached muslin. In these partitioned rooms slept
travellers who arrived at the end of each day (Atwood 1978:80).
Planning for an unknown number of days-end visitors was difficult
but there was always enough food. Sleeping accommodations were
primitive, however, and rest was not always to be easily had.
U.S. Forest Service ranger Henry Haefner recalled a sleepless
stay at the Billings place in the I920's. (Haefner 1975):

... Upstairs was one large room packed full of beds.
Bed sheets suspended from the ceiling by wires
separated some beds from others... Beds were mainly
piles of gunny sacks and old unwashed blankets held up
by a network of rawhide thongs across the underside of
a frame. . When the place was full at night it was a
nightmare. There was almost continuous coughing,
snoring, grinding of teeth, talking in sleep, urinating
in a can or out the window, and other night noises.
There always seemed to be somebody walking around the
room or to the window or stairway, which shook the
floor and building. Sound sleep for any length of time
was impossible.

In 1931 George Billings sold the ranch to Hollywood
businessman Stanley Anderson for $5,000. In later years Anderson
and his sons enlarged the house, added a caretaker's residence,
bunkhouse, tackhouse, woodshed, storage shed, and restored a
blacksmith shop. In 1970 the Anderson family sold the entire
two-hundred acre parcel to the U.S. Government under the National
Wild and Scenic Rivers program.

Small communities remained as important supply centers in
isolated reaches of the canyon until the end of World War II.
Miners constituted most of the customers at Galice, but were
joined by tourists exploring the scenic river area. At Mule
Creek, Marial offered food, lodging, postal service and a river
crossing. The post office opened on January 29, 1903. Marial
Lodge was constructed during the 1920 's to accommodate the area's
ever-growing number of anglers and boaters. With an increase in
mining population and the completion of a road to the mouth of
Mule Creek in the 1930 's more people had access to the supply
center.

121

The U.S. governinent had established its presence in the
Rogue Country in the early twentieth century. With the creation
of the U.S. Forest Service in 1906, management and protection of
lands within the Rogue Canyon became the responsibility of the
new organization. In the fall of that year a Forest Service
office opened in Grants Pass. The Galice District of the new
Forest was administered from Rand, where the District Ranger
Station was constructed in 1917 (Cooper 1939) . By 1920 there were
two Forest Service Guard stations within the river canyon, one at
Whisky Creek and one between Meadow Creek and Horseshoe Bend.
Fire detection and suppression were major tasks assigned to the
Forest Service. Trail construction was also given major emphasis.
Improved trails, telephone lines and fire lookouts were soon
established throughout the canyon area (Beckham 1978:150) .

U.S. Forest Service Crew, ca. 1920

122

Siskiyou National Forest

Following a severe fire season in 1910, the Forest Service
intensified efforts to construct new transportation routes and to
improve old paths through the canyon. Long circuitous trails
over hills and dangerous slide areas were re-routed and made
safe. The Forest Service frequently hired canyon residents and
together they built efficient, dependable travel routes. By 1918
work was underway on a primary route known as the Rogue River
Trail. The trail would serve crews fighting fire, the mail
service, and supply packers into the area. The Gold Beach
newspaper reported:

County Judge W.A. Wood and Forest Supervisor N.F.
Macduff, of Grants Pass, and an engineer of the
Forestry department, went over the plans for the
construction of the trail along Rogue River from Mule
Creek to Kelsey Creek on the county line, the
construction of which will be undertaken as soon as men
and materials can be obtained (Gold Beach Reporter
October 10, 1918) .

Three months later a completion date for the entire trail
was planned:

The Rogue River trail from Almeda to Gold Beach, a
distance of 78 miles, will be completed by July l...The
trail is being built by the Forest Service. Interest is
being shown in the Rogue River trail by anglers of the
state and northwest as that section of the country is
noted for its good fishing (Gold Beach Reporter
December 19, 1918) .

In the 1930 's the Forest Service was assisted in the Rogue
country by Civilian Conservation Corps crews. Under Forest
Service staff supervision additional trails, bridges, and guard
stations were constructed throughout the district. The work of
the Civilian Conservation Corps is addressed in an earlier
section of People and the River.

123

VIII. POST WAR AND RECENT YEARS (1945-1995)

Introduction

After World War II, the Rogue between the Applegate River
and Mule Creek changed in important ways. Upriver, large farms
were divided into smaller parcels and road construction made
transportation easier within the settled areas. In the post-war
years the relationship of people to the land changed as jobs --
many of them timber industry positions -- took residents to
Grants Pass or Medford. By the 1960's some city dwellers
relocated in the rural areas, seeking a respite from the noise
and congestion of larger communities.

In the Rogue Canyon, Civilian Conservation Corps roads
allowed new access to previously isolated areas and trucks
replaced the pack trains. Only a few miners returned to the
streams. The few men who did return occupied old cabins or built
new ones. Joe Utassey, for example, who lived first in a cave
shelter he had dug, moved into a cabin at China Gulch in 1946.
Utassey lived into his eighties in the canyon. Other post-war
river canyon residents included Robert Fox at Battle Bar, Bill
Brockman and Jack Mahoney (Atwood 1978:207) .

At Galice the number of miners decreased and its function as
a center for this widely scattered population declined. The
community gradually developed into a supply point for the
fishing, boating, and rafting groups which arrived through much
of the year. At Marial, older members of the large Billings clan
died in the years following the war. The post office at Marial
closed in 1954. Now anglers, boaters and hikers visited the
canyon each season and lodges flourished at 31ack Bar and Marial.

Opening the River

In the years before the advent of power boats, some
excellent boatmen carried passengers and freight over the
difficult rapids and tumultuous water. Bill Winsor, A.W. Presley,
Henry Moore, Amaziah Aubery, Isaac Fry, Claude Bardon, Frank
Lowery and Glen Wooldridge distinguished themselves as skilled
boat operators. As the number of tourists increased interest
rose in making river passage easier.

Canyon residents involved in attempts to transport equipment
and supplies by boat had long realized that rocks were serious
deterrents to predictable use of the river for freighting
purposes. Early in the century mining equipment was boated and
winched downriver to hydraulic and lode mining sites. As early
as 1906 the Rogue River Courier reported that two boats had
brought a heavy load down to a placer mine at Russian Charley
Bar:

124

...the boat was an ordinary Rogue River fishing boat,
The load consisted of tools for a blacksmith shop,
rubber boots, clothing, drygoods and camp supplies all
weighing about a ton. There was also 200 feet of
lumber . . .

The article further quoted one hopeful canyon resident as
saying, "At small expense the boulders could be blasted out and
the channel cleared so loaded boats could go to any stage of the
water except in very high floods to all parties on the lower
river. "

Of the many who worked the Rogue River, Glen Wooldridge
combined his love of the water, skill at boatbuilding, and the
use of dynamite to open the Rogue to traffic.

Glen Wooldridge

125

Young Wooldridge built his first boat with a friend in
August, 1915 and descended the Rogue to the coast in the newly
constructed craft. Within a short time he decided to build up a
guide business, bringing tourists to the popular river.

Through the 1920' s and 1930 's an increasing number of
anglers and vacationers used Glen Wooldridge' s guide service, and
through these years Wooldridge regularly blasted difficult rocks
from his path in the river. In May, 1947, drawing wide publicity
for the Rogue, Glen Wooldridge began a motorized trip by boat up
the Rogue River from Gold Beach. In the years following the war,
Wooldridge' s blasting activities, which had extended over
decades, increased and the river was fully opened to commercial
fishing and raft trips. He remembered (Atwood 1978) :

We used Blossom Bar for several years before we blasted
it out. We'd never had any experience with powder.
Didn't know how to go about it. Just had to learn. We
shot it out before we ever tried to go up river. . . Any
place of consequence that gave us any trouble at all we
blasted the rocks out.

Blossom Bar

126

We done a lot of work at Kelsey Falls. We changed the
channel from one side of the river to the other... Then
we came on up to Horseshoe Bend and Jenny creek. . .
bombed hell out of that. Then at Upper Black Bar we
opened that channel around next to the bluff. Above
there we opened Wildcat. Done some shooting over on
Tyee, over on the north side. Grave Creek Falls had
some boulders we boosted out... The blasting was
carried on over a long period of years. . . .

The river guide service, of which Glen Wooldridge was a
pioneer, has expanded through the years with a steady increase in
the number of people seeking an opportunity to float or boat the
Rogue. Glen Wooldridge' s long career as well as his efforts at
blasting boulders from the water, had a peirmanent impact on the
river .

Flood and Storm

On several occasions in the past fifty years severe flooding
and winds as well as dynamite have changed the river canyon.
Just before Christmas, 1955, heavy winds and rains lashed the
Pacific Coast region. Lowlands along the Rogue west of Grants
Pass flooded. Eyewitnesses in the Rogue Canyon observed
driftwood along the river high on the banks where the flood had
deposited it in trees. The force of the water wedged logs into
rocks along high cliffs (Pierce 1962:43).

In 1962 the Columbus Day storm brought tremendous winds to
the Rogue Canyon. Dr. and Mrs. Allen Boyden of Portland were
visiting their river canyon property when the storm hit. One
writer described their experience:

The Allen Boydens of Portland, who own a cabin on
Horseshoe Bend, were staying in it at the time of the
storm of October 12th [1962] . They watched from a clear
field while whole groves of virgin timber were
flattened or snapped off to the accompaniment of a
deafening roar... Several times the river seemed to
lift right out of its channel. Tons and tons of water
were carried up into the air and then dropped. The same
phenomenon was observed also at Winkle Bar (Pierce
1962:43) .

Waters rose quickly on the Rogue in December, 19 64. Several
towns along the river were evacuated, and great property damage
was reported in many areas. Ten years later, in January, 1974,
waters rose mid-month, again resulting in severe flooding and
damage .

Flows were high enough during these major flood years to
destroy bridges, roads, built improvements and to inundate

127

agricultural lands and stream courses. The completion of dams at
Lost Creek on the upper Rogue River and on the upper Applegate
River have greatly reduced the chances of major floods on the
middle and lower portions of the Rogue. While protecting soil and
improvements, the dams have also dramatically altered the river's
ecological balance. Benefits of flooding, including renovation of
riparian ecosystems through renewed spawning beds and sorted
gravel sands and fines, no longer occur.

Government Management of the River

The Bureau of Land Management, established in 194 6 out of
the Department of the Interior, took over administration of non-
National Forest timber lands within the watershed in the late
1940' s. The agency also assumed responsibility for managing the
river corridor. The Rogue River Recreation Withdrawal PLO 1726
of September 3, 1958 -- amended in 1959 and 1963 -- reserved all
BLM administered land within one-half mile of the river for the
protection and preservation of the scenic and recreation areas
adjacent to the river.

BLM Staff with Recreation Sign, ca. 1960

128

After World War II, settlers appeared along the Rogue in new
form. These newcomers were primarily part-time recreational users
who occupied existing cabins or built simple dwellings on mining
claims. Most of these structures were used as vacation homes or
adjunct structures for river guide operations.

The Wild and Scenic River Act was passed in 1968 and the
Rogue was subsequently listed. The Bureau of Land Management
assumed responsibility for the river corridor. Intent upon
restoring the canyon to natural conditions, the agency began a
program of identifying mining claims and cabins illegally held.
Cited under "illegal trespass, " title to these lands was
purchased or claimed from owners. With increasing efforts to
protect scenic and natural qualities of the Rogue River Canyon,
the Bureau of Land Management struggled with the long-standing
problem of residential construction on mining claims.

In 1965 the Bureau of Land Management compiled a detailed
report of the "occupancy problem" as they understood it. The
report summarized the status of residential occupancy on the
river in that year. Between Hellgate and Grave Creek the agency
found six full-time occupants and six part-time occupants. Three
residences had been abandoned, with the cases not closed; eleven
cases had been successfully resolved ("Rogue River Occupancy
Problem" 1965:11,9) .

Cabin in Rogue River Canyon

129

Between Grave Creek and Marial the report noted four full-
time residential occupancies -- four full-time and eight part-
time. None of these cases had been closed. The BLM further
found that all of the occupancies were involved with mining
claims and, "in almost all cases the full-time occupants' use of
the land predates the enactment of P.L. 87-851, are elderly-
persons, and make this their principal place of residence.

Recognizing the complexity as well as the time-consuming
aspects of resolving these cases, the BLM planned several
actions. The first two steps involved (1) allowing no new
occupancy or private use on withdrawn lands, and (2) a gradual
reduction in number of existing residential occupancy by issuing
life- time leases where appropriate, or issuing short-term non-
renewable leases to allow occupants time to re-locate. The third
step ordered immediate cancellation of the residency if the
occupancy had occurred after October 23, 1962, if the occupant
refused to comply with terms of the lease, if the occupant had
died, or if the land was needed for a definite project.
Throughout the canyon from Grave Creek to Marial, mining cabins
and "illegal" recreational dwellings were torn down or burned.

In 1995, thirty years after this report was written, few, if
any, of these structures left along the river. As described by
Dennis Gray in the "Rogue River Cultural Resource Inventory, "
(1995) agency staff thoroughly cleared the property:

In those cases where the claim was evaluated as
invalid, the improvements on the property were removed.
The procedure followed by the BLM to remove a structure
(after notification that a claim was null and void) was
to remove appliances (stoves, water heaters, etc.) and
take them to a land- fill, remove the door and window
casings, and then pull down the structure. Finally they
burned the remains in situ (Gray 1995:8).

The removal of these structures within the Hellgate
Recreational and Wild Section of the Rogue River signaled the
beginning of a new era for the Rogue River under government
management. While private property still exists in substantial
areas within the Recreational Section and at a few locations
within the Wild Section, approximately 99% of the land within the
Scenic Corridor is now publicly owned (or has a scenic easement
held) . Responsible agencies include the Bureau of Land
Management, the State of Oregon, and Josephine County (Gray
1995:9) .

With passage of the Wild and Scenic River Act in 1968 and
subsequent listing of the Rogue, public awareness and popularity
of the river grew rapidly. The Wild and Scenic River designation
of the river extends between the mouth of the Applegate and
Marial, on Mule Creek in Curry County. This portion of the river

130

corridor was placed under the jurisdiction of the Bureau of Land
Management. The segment is divided into two designations for
management purposes; Recreational and Wild. The Recreational
section of the Rogue extends from the mouth of the Applegate
River to Grave Creek. The Wild section begins at Grave Creek and
continues to Watson Creek, a distance of thirty-two miles . (Bureau
of Land Management responsibility ends at Mule Creek.)

Following Wild and Scenic River designation, visitors
flocked to the Rogue Canyon and by the mid-1970' s more than
10,000 people per year were using this isolated portion for
boating, hiking, fishing and other recreational pursuits (Purdom
1977:8) . In subsequent years the river has seen increased
recreational usage from a growing local population as well as
tourists from other areas. The Rogue's national reputation as an
excellent salmon and steelhead fishing stream, as well as a river
with outstanding white water boating and rafting opportunities,
brings thousands of visitors each year. Fishing, float trips,
camping, hiking, picnicking, swimming and sightseeing are all
enjoyed within an area of great natural beauty.

fe

C4^f%:^

^^

^^-;^''\z'S-T-^,

H^^^

'^^^^^ ''. *^* ^ ^ '',S'"

P

^ ' .?*<. ^f "j^

'^m

BLM Staff with Rogue River Trail Sign

131

The Bureau of Land Management, with local headquarters at
Rand, today devotes extensive study and staff to continual
stewardship of river, monitoring visitors, size of parties, use
of camping and sanitary facilities, safety, and vandalism in the
Rogue Canyon .

In the 150 years since ending of Native culture and the
present day, the study area along the Rogue has changed in many
ways. In 1845 Indians still fished for salmon on the Rogue.
Today shrieking tourists float the river in rubber rafts.

For thousands of years prior to 1820 native peoples adapted
successfully to their environment. With Euro-American exploration
came an ever- increasing population and advancing technology. The
village sites, fishing locations and food-gathering grounds in
valley areas along the river downstream from the mouth of the
Applegate were transformed into agricultural fields, roads and
built improvements. In the river canyon sites used by Indians
for subsistence activities, were excavated in a relentless search
for gold. Cabins, irrigation ditches and flumes soon crossed the
creeks and bars along the river's edge.

In all instances, bars and terraces and meadows along the
Rogue were usurped and developed to new purposes. Change came
steadily to the river throughout the decades. Mining,
agriculture, roads and ferries, floods, and recreation,
permanently altered the river landscape. From the mouth of the
Applegate to Pickett Creek people still occupy the areas once
inhabited by natives. Within the Wild Section of the Rogue
Canyon, humans continue to use the river, but in a substantially
new way. The gold miners, farmers, and packers are gone. Under
U.S. government management the river canyon is now briefly home
to the large numbers of rafters, anglers, and hikers who visit
the area .

The Rogue River's story has evolved over thousands of years.
The qualities that drew people to the area centuries ago continue
to attract. The gentle climate, sheltering mountains and
isolation historically suited river corridor inhabitants and
entice visitors today. Although the modern period reflects only
a moment in history, the ways in which we use and care for the
Rogue will detennine what remains for future generations who are
drawn to its banks for sustenance or seclusion.

132

Abbett,
1973

Aikens,
1993

Atwood,
1978

Atwood,
1984

Atwood,
1990

BIBLIOGRAPHY

Thomas

Nomination to the National Register of Historic Places,
Whisky Creek Cabin.

Melvin C.

Archaeology of Oregon. U. S. Department of Interior,
Bureau of Land Management, Portland, Oregon.

Kay

Illahe: The Story of Settlement in the Rogue River

Canyon . Ashland, Oregon: Kay Atwood

Kay

Josephine County Historical Resources Inventory. Grants
Pass, Oregon: Josephine County Planning Department.

Kay

Nomination to the National Register of Historic Places,
"Speed's Place on the Rogue."

Atwood,
1983

Kay

Nomination to the National Register of Historic Places:
William H. and Viola Flanagan House, Grants Pass,
Oregon .

Atwood,
1976

Kay

Oral History Review: Rogue River. Transcripts of
oral interviews with residents of the Rogue River
Canyon. Manuscript and tape recordings on file with
the Oregon Historical Society, Portland, Oregon.
Interviews conducted with George Morey, Glen
Wooldridge, Red Keller, Lou Martin and
Ivin Billings.

Barnett, Homer G.

1937 Culture Element Distributions: VII Oregon Coast,

Anthropological Records. Vol. 1., Number 3:155-203.
University of California Press. Berkeley, California.

Beckham, Steven Dow

1993 Takelman and Athapascan Lifeways and History, Rogue
River Corridor- -Applegate River to Grave Creek:
Investigations for Interpretive Program. Report to the
U.S .D. I . -Bureau of Land Management, Medford District.
Heritage Research Associates, Inc., Eugene, Oregon.

133

Beckham, Steven Dow

1978 Cultural Resource Overview of the Siskiyou National
Forest. U.S.D.A. Forest Service, Siskiyou National
Forest, Grants Pass, Oregon.

Beckham, Stephen Dow

19 71 Requiem For A People: The Rogue Indians and the

Frontiersmen. Norman, Oklahoma: University of Oklahoma
Press .

Brauner, David

1983 The Reevaluation of Cultural Resources within the

Applegate Lake Project Area, Jackson County, Oregon,
Phase III: Data Recovery of Site 35JA42 . Report to
U.S. Army Corps of Engineers. Department of
Anthropology, Oregon State University, Corvallis,
Oregon.

Brooks, Howard; Ramp, Len

1968 Gold and Silver in Oregon. Bulletin 61. Salem, Oregon:
State Department of Geology and Mineral Industries.

Bureau of Land Management-USDI/Forest Service-USDA

1993 Rogue River Float Guide . Published by the USDA-Forest
Service and the USDI -Bureau of Land Management.

Bureau of Land Management

1965-79 Occupancy Trespass Files. Information on file with the
Medford District of the Bureau of Land Management,
Medford, Oregon.

Cassidy, Julie Krieger

1992 Explaining Lithic Assemblage Differences in the

Medicine Lake Highlands, Siskiyou County, California.
U.S. Department of Agriculture, Forest Service.
Redding, California.

Clark, Linda

1988 A Study of Temporal Variation in the Scrapers from the
Marial Site (35CU84) . Report prepared for the Medford
District of the Bureau of Land Management, Medford,
Oregon.

Connolly, Thomas

1990 Cultural Stability and Change in Southwest Oregon and
Northern California: An Approach to Identifying
Diagnostic Assemblage Types. In, Living with the Land,
edited by Nan Hannon and Rich Olmo, pp 56-62. Southern
Oregon Historical Society, Medford, Oregon.

134

Connolly, Thomas

1986 Cultural Stability and Change in the Prehistory of
Southwest Oregon and Northern California. Ph.D.
dissertation, Department of Anthropology, University
of Oregon, Eugene, Oregon.

Cooper, Loren

1939 A History of the Siskiyou National Forest, U.S.
Service .

Forest

Deich, Lyman

1983 35CU84 (Rogue River Ranch) A Pre- Preliminary Report of
Excavations Carried Out During the 19 82 Field Season.
Paper presented at the State of Jefferson Conference,
February 1983. On file, Medford District Office of the
Bureau of Land Management, Medford, Oregon

Deich, Lyman

1976-78 Cultural Resource Inventory Forms. Site records on
file with the Medford District, Bureau of Land
Management, Medford, Oregon.

Diller, J.S.

1914 Mineral Resources of Southwestern Oregon. Bulletin 546.
Washington: Department of the Interior, United States
Geological Survey.

DOGAMI
1952

Oregon Metal Mines Handbook. Bulletin 14-C, Volume II
Salem, Oregon: State of Oregon Department of Geology
and Mineral Industries.

Dorsey, J. Owen

1890 The Gentile System of the Siletz Tribes, University of
California Publications in American Archaeology and
Ethnology. Volume 36:221-300. Berkeley, California.

Drucker, Philip

1940 The Tolowa and Their Southwest Oregon Kin.

University of California Publications in American
Archaeology and Ethnology. Vol. 36: 221-300.
Berkeley, California.

Eastwick, Philip

1879 U.S. Corps of Engineers, Annual Report of the Chief
Engineers to the Secretary of War for the Year 1879,
Part II. Washington D.C.: Government Printing Office.

Follansbee, Julia A., Pollock, Nancy L.

1978 Prehistory and History of the Jackson-Klamath Planning
Unit: A Cultural Resources Overview. Bureau of Land
Management

135

1856 Galice Store Records 1856-1857. Josephine County
Historical Society, Grants Pass, Oregon.

General Land Office (GLO)

1856- General Land Office Maps (various dates) on file
Siskiyou National Forest, Grants Pass, Oregon.

Gray, Dennis J.

1995 Rogue River Cultural Resource Inventory. Report
prepared for the Medford District, Bureau of Land
Management, Medford, Oregon.

Gray, Dennis J.

1994 Rogue River Survey Literature Review. Report prepared
for the Medford District, Bureau of Land Management,
Medford, Oregon.

Gray, Dennis J.

1987 The Takelma and Their Athapascan Neighbors. University
of Oregon Anthropological Papers No. 37, University of
Oregon, Eugene, Oregon.

Grey, Zane

1928 Tales of Freshwater Fishing. New York: Harper &
Brothers .

Haefner, Henry E.

1975 Reminiscences of an Early Forester. Oregon Historical
Quarterly Vol . LXXVI .

Hannon, Nan

1992 Hand to Mouth, Plant Food Resources and Prehistoric
People in Southwest Oregon's Bear -Creek Valley.
Unpublished M.A. thesis. Department of
Sociology/Anthropology, Southern Oregon State College,
Ashland, Oregon.

Harrington, John Peabody

1981 The Papers of John Peabody Harrington in the National
Anthropological Archives of the Smithsonian
Institution. 1907-1957. Reel Number 28. Kraus
International Publications. Millwood, New York.

Highsmith, Richard M. and A. Jon Kimerling

1979 Atlas of the Pacific Northwest. Oregon State
University Press, Corvallis, Oregon.

Hill, Edna May.

1976 Josephine County Historical Highlights. Grants Pass,
Oregon: Josephine County Library System; Josephine
County Historical Society.

136

Holt, Catharine

1946 Shasta Ethnography. University of California
Anthropolocrical Records. 3:4:299-346.

Jacobs, Melville

n.d. Galice Creek Field Notes of Melville Jacobs. Box 104,
Number 1693-71-13 of the Melville Jacobs Collection.
Archives of the University of Washington, Seattle,
Washington.

Korbulic, Mary.

1992 "Zane Grey: Romancing the Rogue," Table Rock Sentinel
(November-December) .

Kroeber, A. L.

1953 Handbook of the Indians of California. California Book
Company, Ltd. Berkeley, California. (First published
in 1925 by the Bureau of American Ethnology.)

LaLande, Jeffrey

19 87 First Over the Siskiyous: Peter Skene Qaden's 1826-1827
Journey through the Oregon-California Borderlands,
Portland, Oregon: Oregon Historical Society.

LaLande, Jeffrey

1990 The Indians of Southwestern Oregon: An Ethnohistorical
Review. In Living with the Land edited by Nan Hannon
and Rich Olmo, pp 95-119. Southern Oregon Historical
Society, Medford, Oregon.

LaLande , Je f f rey

1980 Prehistory and History of the Rogue River National

Forest: A Cultural Resource Overview. Medford, Oregon:
U.S. Forest Service, Rogue River National Forest.

Libby, F.W.

1967 The Almeda Mine of Josephine County. Oregon. Salem,
Oregon: State of Oregon Department of Geology and
Mineral Industries.

McArthur, Lewis L.

1992 Oregon Geographic Names. Portland, Oregon: Oregon
Historical Society.

Mickey, John

1859 Diary. MS in possession of U.S. Department of the

Interior, Bureau of Land Management, Medford, Oregon.

Morey, George

n.d. George Morey Shares Recollections About CCC Camps
during 1930' s. Medford, Oregon: Bureau of Land
Management

137

Nilsson, Elena and Michael Kelly

1991 Prehistory of the Upper Rogue River Region:
Archaeological Inventory and Evaluation within the
Elk Creek and Lost Creek Lake Project Areas, Jackson
County, Southwest Oregon. Report prepared for the
U.S. Army of Engineers, Portland District, Portland,
Oregon.

O'Neill, Brian L.

1992 Pre-Mazama Occupation of the Dry Creek Site (35DO401) ,
Southwest Oregon. Paper presented at the 45th annual
Northwest Anthropological Conference, Simon Eraser
University, Burnaby, British Columbia, Canada.

Ottis, Nancy

1991 Archaeological Evaluation of the Stratton Creek Site
{35J021) . Report prepared for the Bureau of Land
Management, Medford District, Medford, Oregon.

Pettigrew, Richard M.

1990 New Pathways for Research in Southwest Oregon

Archaeology. In, Living with the Land, edited by Nan
Hannon and Rich Olmo, pp 63-69. Southern Oregon
Historical Society, Medford, Oregon.

Pettigrew, Richard and Clayton Lebow

1987 Data Recovery at Sites 35JA27. 35JA59. and 35JA100.

Elk Creek Lake Project, Jackson County, Oregon. Report
by INFOTEC Research Inc., Eugene, Oregon for the U.S.
Army Corps of Engineers, Portland, Oregon.

Pierce, Bob

19 62 Early Days on the Rogue, Mazama vol. 44, No. 13.

Poesch, Jessie

19 62 Titian Ramsay Peale 1799-1885 And His Journals of the
Wilkes Expedition. Philadelphia: The American
Philosophical Society.

Price, Henry Teller

1967 Up the Rogue River and the First Mail Route. Portland,
Oregon: Metropolitan Press.

Purdom, William

19 7 7 Guide to the Geology and Lore of the Wild Reach of the
Rogue River Oregon. Bulletin No 22. Eugene, Oregon:
University of Oregon Museum of Natural History.

Ross, Richard

1987 Marial . Research proposal submitted to the Bureau of
Land Management, Medford District, by Oregon State
University.

138

Ross, Richard and JoAnn Blalack

1994 Stratton Creek (35J021) Archaeological Investigations
1992 . Report prepared for the Bureau of Land
Management, Medford District, Medford, Oregon.

Sapir, Edward

1907 Notes on the Takelma Indians of Southwestern Oregon,
American Anthropologist. Volume 9, Number 2:251-275.

Schreindorf er, Crystal S.

1985 Marial : 1984, Archaeological Investigations at 35CU84.
Report prepared for the Bureau of Land Management .
Department of Anthropology, Oregon State University,
Corvallis, Oregon.

Spencer, Lee

1991 Pacific Salmon and Environmental Change in the Rogue
Basin. Report prepared for the Swedenburg Cultural
Resource Center, Ashland, Oregon.

Steele, Harvey

1984 The Marthaller Site. Report No. 9, Oregon

Archaeological Society. Report on file at the Medford
District of the Bureau of Land Management, Medford,
Oregon .

U.S. Bureau of Census

1870 Josephine County, Oregon Census Records for 1870.

U.S. D.I. Bureau of Land Management

1965 Rogue River Occupancy Problem. Part II.

Walling, A.G.

1884 A History of Southern Oregon Comprising Jackson,

Josephine, Douglas, Curry and Coos Counties. Portland,
Oregon: A.G. Walling.

Wallmann, Steve

1994 Interview conducted 1/19/94 by Dennis Gray, Cascade
Research, Ashland, Oregon.

Wilson, Bart

1979 Excavations at the Ritsch Site. Master's thesis,
Oregon State University, Coirvallis, Oregon.

Winthrop, Kathryn
1993 Prehistoric Settlement Patterns in Southwest Oregon.
Ph.D dissertation. Department of Anthropology,
University of Oregon, Eugene, Oregon,

139

Winthrop, Kathryn

1993a Cultural Resource Inventory Forms. Site records on
file with the Medford District, Bureau of Land
Management, Medford, Oregon.

Winthrop Kate

1995 "Cultural Resources Background Paper for Revising the
Hellgate Recreation Area Management Plan." U.S. D.I.
Bureau of Land Management, Medford District Office.

Newspapers

Crescent City Herald. March 24, 1858.

Democratic Times. (Jacksonville, Oregon) February 6, 1890.

Gold Beach Reporter, December 19, 1918.

Grants Pass Courier, March 4, 1927.

Ivin Billings Interview. Rogue Coast (Gold Beach Oregon) , August
29, 1974.

Oregon Sentinel Jacksonville, Oregon December 31, 1861.

Port Orford Post. March 17, 1881.

Portland Oregonian. July 3, 1852.

Rogue River Courier. January 29, 1903.

Twenty-Five Miles of New Road Built. Grants Pass Courier,
September 1, 1934.

140

S. GOVERNMENT PRINTING OFFICE; 1996 - 790-118 / 21219 REGION NO. 10

t 882 .R6 A89 1995 v. 1

Atwood, Kay.

People and the river

u

SLM LIBRARY

'E'VVER FEDERAL CENTER
^^ PO BOX 25047 ^
DENVER, CO 80225

w

oo

O JD

^§

m

go

<=^

en o

H 3)

m m
o CO

0)

m

oQ

33>

5i

- C/J

u
o
o

Q. 00

o g

a o

5 1

CD Q.
eg (D

3 J3
o

CO

en
o

Q}

J3

m

>

s

c

CD

a.

o

o

■n

a

I-

D

>

z

w"

a

— ^

o'

s

O

>

z

ZT:

>

O
CD

o

m

m

D
m

>

u

■HOT

m

zii
-^0
3:1